icon-cookie
The website uses cookies to optimize your user experience. Using this website grants us the permission to collect certain information essential to the provision of our services to you, but you may change the cookie settings within your browser any time you wish. Learn more
I agree
richard yuwen
305 articles
My Web Markups - richard yuwen
  • 产生了事件
  • 用户执行了命令
  • DDD 进行系统分析方面有了一个新兴的方法论，即 Event Storming(事件风暴)
DDD 实践手册(番外篇: 事件风暴-概念) - InfoQ 写作平台
3 annotations
 
xie.infoq.cn
633
  • Query 端使用与查询需求更为贴近的数据存储引擎
  • 将系统所有产生业务行为以 append-only 的形式存储起来
  • 使用领域驱动的思路设计查询接口
  • 每一个查询接口设计一个数据视图
  • 查询功能的数据模型
  • 可靠的消息中间件
  • 业务对于数据完整的实时性非常高，那么可能 CQRS 不一定适合你
  • 关系型数据库的事务特性
  • 「领域事件」的方式通知 query 系统
  • CQRS 的核心思想是将这两类不同的操作进行分离
  • Command Query Responsibility Segregation
  • 分层的方式将业务逻辑剥离出来，单独进行维护，从而控制业务本身的复杂度
  • DDD 通过领域对象之间的交互实现业务逻辑与流程
DDD 中的那些模式 — CQRS - InfoQ 写作平台
13 annotations
 
xie.infoq.cn
545
  • Bookmark 特性
  • Priority and Fairness 特性
  • 最大读和写并发数
  • 处理已取消请求的 goruntine
  • NodeLease 对象 (0.1 KB) 更新请求替换老的 Update Node Status
  • 先扩后缩
  • 优化客户端将请求平衡掉来解决
  • 组件对 API Server 的请求链路健康非常敏感
  • API Server 的资源优先级设置到最高级别
  • NodeLease
  • 瓶颈在 Kubernetes API Server
  • 大规模集群维持在健康状态需要全链路优化和调优
  • 压测场景其实非常片面
  • 差异化部署
  • 联合的、可自动生成的生命周期框架
  • 声明式基础设施
  • Kubernetes 发布还是不够透明
  • 全新的沙箱集群
  • 镜像化
  • 沙箱集群测试
  • 自动化和 CI/CD 是云原生非常重视的理念
  • Node-Operator 接管了几乎整个蚂蚁金服的物理机生命周期
  • Node Problem Detector） 上报的节点故障信息
  • 面向终态的特性去管理 Worker 节点
  • 声明式编程
  • 自动化维护 Worker 节点
  • 分钟级甚至秒级创建一个 Kubernetes 集群
  • 元集群
  • 业务集群
  • 交付过程封装成 Operator
  • Kubernetes 运行在 Kubernetes 之上
  • 无差别运行 Kubernetes
  • Kubernetes 云原生化
  • 过程自动化
  • 云原生技术让开发和迭代敏捷化
  • 功能对齐
一年时间打造全球最大规模之一的Kubernetes集群，蚂蚁金服怎么做到的？ - InfoQ
36 annotations
 
www.infoq.cn
607
  • generic controllers
  • planning and design stage of building a web framework
Build a web framework · Build web application with Golang
2 annotations
 
astaxie.gitbooks.io
521
  • RESTful controllers
  • flow of the application
  • Anything you intend to do well must first be planned well
Project program · Build web application with Golang
3 annotations
 
astaxie.gitbooks.io
584
混布评估后台 - Feishu Docs
0 annotations
 
bytedance.feishu.cn
589
  • 配置文件的格式，我个人使用 .json 或 .yaml
使用 Iris 打造一款个人博客(配置篇) - 知乎
1 annotations
 
zhuanlan.zhihu.com
529
  • 21st July
Awesome Kubernetes
1 annotations
 
ramitsurana.github.io
575
  • 集中的策略管理和实施
  • 策略管理
Kubernetes治理，你应该知道的 - 云+社区 - 腾讯云
2 annotations
 
cloud.tencent.com
542
选择 AirPods Pro 耳塞并进行耳塞贴合度测试 - Apple 支持
0 annotations
 
support.apple.com
511
airpods怎么查激活日期_查询序列号_智能家
0 annotations
 
www.znj.com
427
  • Built-in Func
  • testInput
  • Party.ConfigureContainer
  • a function dependency can accept previous registered dependencies and update or return a new value of any type
  • Basic Authentication
  • Cookies
  • Model binding and validation
  • Using middleware
  • Grouping routes
  • Querystring parameters
  • Register Dependencies
  • Custom Preflight
  • Request & Response & Path Parameters
  • hero
  • Dependency Injection
  • Parameters in path
Start - Iris Web Framework
16 annotations
 
iris-go.com
451
  • 领域层有着明确的 BC 划分
  • 粗粒度」业务的入口
  • Use Case
  • Data Transfer Object
  • Bounded Context(
  • BusinessException
  • 充血模型
  • 领域对象
  • 仓库
  • 领域服务
DDD 实践手册(2. 实现分层架构) - InfoQ 写作平台
10 annotations
 
xie.infoq.cn
470
  • truly safe bindings.
dependency injection · kataras/iris Wiki
1 annotations
 
github.com
370
Installation · kataras/iris Wiki
0 annotations
 
github.com
380
  • effect of the panic
  • A Handler, as the name implies, handle requests.
Routing · kataras/iris Wiki
2 annotations
 
github.com
341
  • WithConfiguration
iris - GoDoc
1 annotations
 
godoc.org
312
  • Third-Party
  • Party
  • Community
  • Built-in
  • Built
  • free to use any third-party middleware that is compatible with the net/http package
Routing middleware · kataras/iris Wiki
6 annotations
 
github.com
306
  • MVC Controllers are responsible for controlling the flow of the application execution
iris/_examples/mvc/overview at master · kataras/iris
1 annotations
 
github.com
294
  • Nvidia 的领域知识
  • GPU 的花费则是从单 GPU 每小时 10 元 ~ 30 元不等
  • CPU 通常是一小时几毛钱
  • 深度学习所依赖的异构设备及英伟达的 GPU 支持
  • Kubernetes 管理 GPU 能带来什么好处呢
  • GPU 为代表的异构资源
  • GPU 的容器化
  • Kubernetes 是容器调度平台
  • 调度单元是容器
  • 构建支持 GPU 的容器镜像；
  • 利用 Docker 将该镜像运行起来
  • GPU 设备和依赖库映射到容器中
  • Nvidia 硬件驱动
  • 在宿主机上安装 Nvidia 驱动
  • 而在 CUDA 以上的软件交给容器镜像来做
  • Mount Bind 的方式映射到容器中
  • 通常大家会使用 Nvidia-docker 来运行 GPU 容器，而 Nvidia-docker 的实际工作就是来自动化做这两个工作
  • GPU 应用依赖的驱动库
  • Nvidia 驱动
  • Nvidia Docker2
  • Nvidia Device Plugin
  • 节点的状态信息中发现相关的 GPU 信息
  • limit 字段中指定 nvidia.com/gpu 使用 GPU 的数量
  • nvidia-smi
  • 启动时刻的资源上报
  • 用户使用时刻的调度和运行
  • Allocate
  • ListAndWatch
从零开始入门 K8s | GPU 管理和 Device Plugin 工作机制 - 阿里云云栖社区 - 博客园
28 annotations
 
www.cnblogs.com
90
  • /var/lib/kubelet/device-plugins/
  • vendor-domain/vendor-device
  • vendor independent solution
  • consistent and portable solution
  • monitor them
  • vendors
  • advertise their resources
  • Device Manager Proposal
community/device-plugin.md at master · kubernetes/community
8 annotations
 
github.com
74
  • DAO 在实际项目中往往会更底层些，它抽象的是不同关系型数据库的异同
  • 领域对象对应的数据存储起来
  • 领域服务就对应了动词
  • 领域对象从某种程度上代表了领域知识中的名词
  • 无法表达业务含义
  • 函数重载(Overload)
  • 数据与方法
  • 「Effective Java」中的开篇就建议大家尽量使用 Factory 工厂模式来创建对象
  • 领域的对象的核心概念
DDD 实践手册(5. Factory 与 Repository) - InfoQ 写作平台
9 annotations
 
xie.infoq.cn
633
  • Adapter 将输入的数据适配为 BC 内部服务所需的数据对象
  • 「防腐层」对外提供了基于 Facade 模式的粗粒度接口
  • BC 的划分不仅需要系统分析与架构的知识
  • 软件架构往往受到「康威定理」的影响
  • 从业务的角度出发，对系统进行某种维度的拆分
  • 系统代码不仅应该在架构层次进行划分，以区分不同抽象层次以及功
DDD 实践手册(6. Bounded Context - 限界上下文) - InfoQ 写作平台
6 annotations
 
xie.infoq.cn
657
  • 每个实例可以维护一份被返回对象的独立的拷贝
  • 使用 $emit(eventName) 触发事件
  • 使用 $on(eventName) 监听事件
  • props 指定验证要求
  • v-bind 绑定 HTML 特性到一个表达式
  • 接受父组件传递过来的数据的一个自定义属性
  • 实例选项中注册局部组件
Vue.js 组件 | 菜鸟教程
7 annotations
 
www.runoob.com
577
  • vue-webpack
  • 前端资源模块化管理和打包工具
  • 很多松散的模块按照依赖和规则打包成符合生产环境部署的前端资源
  • 按需加载的模块进行代码分割
  • 实际需要的时候再异步加载
  • 通过 loader 的转换，任何形式的资源都可以视作模块
  • CommonJs 模块
  • AMD 模块、
  • CSS、图片、 JSON、Coffeescript、 LESS
  • ES6 模块
[SpringBoot+VueJs] 3.1 VueJS 2.0 + Webpack工程介绍 - 简书
10 annotations
 
www.jianshu.com
505
  • 领域事件的方式将单个 Aggregate 更新的事件广播出去
  • 合适的 Aggregate Root
  • 粗粒度的方法，将实现的逻辑隐藏在这个方法之下，而不是暴露给客户端
  • 领域数据进行操作
  • Aggregate Root，一般称之为聚合根
  • Aggregate 是一种模式
DDD 实践手册(4. Aggregate — 聚合) - InfoQ 写作平台
6 annotations
 
xie.infoq.cn
640
  • 只有心态有了从容，面对金钱、名利才能豁达而又不失进取
  • 人生总会有很多的缺憾，从容的心态是人生最高的境界
  • 人生总会有很多的缺憾，从容的心态是
  • 只有经历过无数次的过失与缺憾，才能在风雨之后看到彩虹
  • 不完美绝非自暴自弃，无所事事的弱价值取向
  • 无为而治
  • 人生境界还有一个层次，那就是经过奋斗把不完满变得完满些
  • 只豁达而不奋斗，就成了玩世不恭，那才是彻底的失败者
  • 金无足赤，人无完人
  • 追求完美即是不完美
次优是人生的最高境界_意林杂志_新浪博客
10 annotations
 
blog.sina.com.cn
600
  • 解耦
  • 每个数据表作为一个实体类，实体类封装的属性对应到表中的字段，这样的话，实体在贯穿于三层之间时，就可以实现增删改查数据了
  • 用户在界面上轻松的增、删、改、查，那么数据库中也要有相应的增、删、改、查，
三层架构详解 | 菜鸟教程
3 annotations
 
www.runoob.com
590
  • 领域层对象组装时，需要完整读取所有 PO 的数据
  • Entity，Value Object 等领域层对象与 PO 的转换逻辑
  • PO 实现 Entity 与 Value Object
  • Domain 层的对象应该保持简单
  • PO 的好处在于将 Entity，Value Object 等对象与持久化框架解耦
  • Value Object 顾名思义，关注的是数据
  • Entity 应该具唯一的「标识」
  • 业务需求中找到那些关键的「业务对象」
  • Value Object(值对象)
  • Entity(实体
DDD 实践手册(3. Entity, Value Object) - InfoQ 写作平台
10 annotations
 
xie.infoq.cn
643
  • 评分函数
  • GPU卡的连接拓扑
  • 负载局限在卡上绑定的容器数
  • Allocate的时候调用Redis的信息查看负载情况
  • 外部数据库存放每个节点的GPU负载信息
  • NVML获取GPU卡的实时状态
  • 任务创建时的真实的GPU负载来动态的决定挂载GPU的id
  • 虚拟id到真实id的映射不能够是通过一个简单的线性映射关系
  • GPU负载的大体上的平衡
  • 伪造出更多虚拟的GPU ID给K8s
  • 没有考虑GPU的亲和性，
  • 提供共享机制最大化资源的使用
  • 显存分配的策略
  • 没有任何gpu虚拟化
  • llocate将在容器创建被调用，用来返回一些能够使用主机上该资源的特殊配置，比如一些环境变量，再将这些信息给到Kubelet，并在容器启动的时候传给容器
  • vendor-domain/resource
  • K8s注册该资源
  • Device-Plugin机制本质上是一个RPC服务
  • AMD GPU等
  • RDMA设备
  • NVIDIA_VISIBLE_DEVICES=0,1
  • 环境变量来指定将要挂载的GPU设备
  • 多个Nvidia-Docker可以挂载同一个GPU
  • 主机侧的Nvidia-Driver
  • libnvidia-container暴露的接口进行交互
  • Runc会调用一个叫做nvidia-container-runtime-hook的hook
  • GPU的支持放入一个兼容OCI标准的运行时库拓展libnvidia-container中
  • Containerd则包装了Runc和其它功能如生命周期管理等，以Daemon的形式运行在主机
  • Nvidia-Docker来做AI系统环境已经是非常主流的做法
  • 原生Nvidia-Device-Plugin
  • K8s中将GPU作为拓展资源调度的Device-Plugin机制
  • 容器GPU挂载
  • GPU调度的通用流程
  • 没有考虑GPU卡之间的通道亲和性
  • 每一块GPU同时最多只能被一个容器使用
  • Device-Plugin的方式来增加对默认资源(CPU,Memory等)之外的设备
  • 借助Kubernetes来管理Nvidia-Docker，使得GPU任务的分配更加简单和合理，目前已成为几乎所有主流的AI算力平台的方案
  • 容器粒度来管理和使用GPU要比从主机角度容易很多
  • Nvidia公司为Docker写的Runtime
  • Nvidia的GPU的利用
如何在Kubernetes实现GPU调度及共享_RancherLabs_传送门
40 annotations
 
chuansongme.com
625
  • native GPU support
  • ative GPU support
  • NVIDIA GPUs are now natively supported as devices in the Docker runtime.
  • Docker 19.03
  • NVIDIA driver
  • utilities
  • a container runtime library
  • NVIDIA Container Toolkit
NVIDIA/nvidia-docker: Build and run Docker containers leveraging NVIDIA GPUs
8 annotations
 
github.com
579
  • 可被抢占实例
(3 封私信 / 2 条消息) 有没有性价比高的GPU云服务器？ - 知乎
1 annotations
 
www.zhihu.com
620
  • QoS控制技术保障了每个用户的GPU资源不被其他租户抢占
  • 实现不同用户之间的强隔离，有效防止信息泄漏
  • 安全隔离技术
  • 基于工作负载需要的资源分配
  • KVM架构与NVIDIA合作研发了异构计算产品
  • 深度学习推理场景对GPU的计算资源消耗可能仅仅是训练阶段的数十分之一，甚至更少
  • 计算资源过剩
  • 深度的软硬件结合及优化才能发挥其性能优势
  • 云端异构计算并非单纯地堆积硬件性能
  • 虚拟化环境中更为灵活
  • Turing架构
  • 可加速深度学习
  • NVIDIA Tesla P4 GPU
  • 单颗物理GPU更细粒度的服务
  • 公共云虚拟GPU异构计算产品VGN5i实例
国内首家，阿里云发布基于公共云的虚拟GPU服务 – NVIDIA 中国官方博客
15 annotations
 
blogs.nvidia.cn
606
  • Run GPU sharing workloads with Kubernetes + Kubeflow
  • sharing GPU to multiple containers,
Is sharing GPU to multiple containers feasible? · Issue #52757 · kubernetes/kubernetes
2 annotations
 
github.com
620
  • Nvidia MPS as the option for isolation
  • helm
  • Kubernetes
CNCF-GSoC/README.md at master · Sakuralbj/CNCF-GSoC
3 annotations
 
github.com
575
  • Scheduling
  • Isolation
gpushare-scheduler-extender/designs.md at master · AliyunContainerService/gpushare-scheduler-extender
2 annotations
 
github.com
600
  • 卡米学习英语小说
  • PDF编辑器和注释器
  • 插入文本等功能进行数字阅读学习。
  • 下划线
  • 突出显示
  • Kami允许多人同时共享编辑PDF，图像或其他可打印文档
  • Kami是适用于Chrome的PDF和文档阅读注释类插件
学用系列｜Kami，基于谷歌课堂的阅读教学神器（附应用案例介绍） - 知乎
7 annotations
 
zhuanlan.zhihu.com
590
  • 整洁架
  • 核心是如何管理层与层之间的交互
  • Application 与 Domain 是实现业务规则的核心，它们不应该依赖于某个特殊的框架或是技术
  • 层与层之间的依赖关系是单向的，是自上而下的
  • 每一层只能依赖自己以下那一层的服务
  • Application 与 Domain 两层
  • 复杂性
  • 技术，业务价值之间找到 trade-off
  • 近几年随着微服务的流行，领域驱动设计(Domain-Driven Design) 重新回到了主流视野中
  • DDD 对核心遗留系统进行重构与拆分
  • DDD 核心的知识与概念
DDD 实践手册(1.Get Started) - InfoQ 写作平台
11 annotations
 
xie.infoq.cn
553
  • 数据开发是一整套数据加工以及加工过程管控的工具，有经验的数据开发、算法建模人员利用数据加工模块提供的功能，可以快速把数据加工成对业务有价值的形式，提供给业务使用
  • 业务系统、日志、文件、网络等，这些数据分散在不同的网络环境和存储平台中
  • 入口
  • 数据资产变为数据服务能力
  • 数据的使用成本更低
  • 屏蔽掉底层存储平台的计算技术复杂性
  • 数据中台提供的工具、方法和运行机制，把数据变为一种服务能力，让数据更方便地被业务所使用
一个通用的数据中台架构应该如何构建？ - InfoQ
7 annotations
 
www.infoq.cn
600
平说春秋145.伍子胥之死_平说平论_新浪博客
0 annotations
 
blog.sina.com.cn
635
  • general availability of the StatefulSets
  • ability to increase the size of persistent volume
  • topology-aware volume scheduling
  • packing plan for a topology
  • Heron
  • node affinity feature
  • pods affinity features allow you to place pods relative to one another
  • optimised for big data workloads
  • new Kubernetes native big data frameworks
  • bare-metal
  • container-based
  • BigBench
  • characteristic
  • networking
  • stateful data
  • smart schedulers
  • containerized data workloads
  • go to container orchestration platform
Kubernetes for Big Data Workloads
18 annotations
 
www.abhishek-tiwari.com
583
  • 使创建的应用程序完全不受云环境和依赖关系的影响。通过实现此目标，可以将应用程序部署在不同的虚拟环境中，而无需特定的驱动程序或要求
“12-Factor”，你准备好了吗？ - DockOne.io
1 annotations
 
dockone.io
636
  • integer resources and cannot be overcommitted
  • Topology Manager
  • co-ordintated in a Topology aligned manne
  • monitor resources provided by device plugins
  • pod-resources
  • gRPC service to enable discovery of in-use devices
  • privileged security context
  • re-register itself with the new kubelet instance.
  • detect kubelet restarts
  • passes this information to the container runtime
  • container runtime configurations for accessing the allocated devices
  • Allocate
  • runs in serving mode
  • starts a gRPC service, with a Unix socket under host path /var/lib/kubelet/device-plugins/
  • Initialization
  • cannot be shared among Containers
  • integer resources and cannot be overcommitte
  • kubelet node status update
  • advertising those resources to the API server
  • list of devices it manages
  • ResourceName
  • Device Plugin API version
  • name of its Unix socket
  • register itself with the kubelet through this gRPC service
  • vendors can implement a device plugin that you deploy either manually or as a DaemonSetEnsures a copy of a Pod is running across a set of nodes in a cluster.
Device Plugins | Kubernetes
25 annotations
 
kubernetes.io
616
  • automatically labels your nodes with GPU device properties
  • Automatic node labelling
  • third-party device plugin
  • must be configured as the default runtime for Docker
  • Docker as its container runtime
  • nvidia-docker 2.0
  • NVIDIA drivers
  • It is not possible to request a fraction of a GPU
  • Containers (and Pods) do not share GPUs
  • specified in the limits
  • schedulable resource
  • install GPU drivers
  • run the corresponding device plugin
Schedule GPUs | Kubernetes
13 annotations
 
kubernetes.io
621
  • 覃超大魔王
纪录Vue.js尤雨溪 - 中英CC字幕 - Honeypot呈现_哔哩哔哩 (゜-゜)つロ 干杯~-bilibili
1 annotations
 
www.bilibili.com
644
  • 通过一个名称来标识一个路由显得更方便一些，特别是在链接一个路由
命名路由 | Vue Router
1 annotations
 
router.vuejs.org
255
  • babel将不被支持的import编译为当前受到广泛支持的 require
  • export 关键字导出模块
  • import 关键字引入模块
  • ES6的模块化方案是真正的规范
  • require来加载一个模块
  • define来定义一个模块
  • CommonJS的核心思想就是通过 require 方法来同步加载所要依赖的其他模块，然后通过 exports 或者 module.exports 来导出需要暴露的接口
快速了解JavaScript模块化
7 annotations
 
www.toutiao.com
435
  • Go uses goroutines for every job initiated by Conn in order to achieve high concurrency and performance
Get into http package · Build web application with Golang
1 annotations
 
astaxie.gitbooks.io
576
  • starts failing the health check
  • a probe CLI tool
  • “exec” probe
  • exposes an HTTP endpoint and calls your app
  • sidecar pattern
  • Sidecar health server
  • expose a http port
  • refactor
  • gRPC application
  • Being able to open a socket doesn’t always mean healthy
  • three types of health probes
Advanced Health Check Patterns in Kubernetes
11 annotations
 
ahmet.im
608
  • active-class
  • tag
  • Vue.js + vue-router 可以很简单的实现单页应用
  • <router-link> 是一个组件，该组件用于设置一个导航链接，切换不同 HTML 内容。 to 属性为目标地址， 即要显示的内容
Vue.js 路由 | 菜鸟教程
4 annotations
 
www.runoob.com
443
  • per-second rolled-up time-series
  • tens of thousands of service interactions per second
  • challenge is to understand performance of a large number of “mice” interactions
  • compute bound or disk I/O bound.
  • organizations which design systems … are constrained to produce designs which are copies of the communication structures of these organizations
  • SLA violations
  • performance drifts
  • service topology view is the right abstraction to visualize the complex control flow (and detect hotspots) in microservices.
  • service interaction analytics
  • evolve from being instance-centric to service-centric
  • new “pets”
  • cattle
  • fault correlation and locating hotspots is hard without understanding service-level dependencies
  • alert fatigue
  • Service Topology Abstraction
  • historical metrics
  • underlying real-time
  • render advanced visualizations
  • alert on SLA violations
  • ad-hoc queries to detect anomalies
  • periodic cron job
  • continuous queries
  • low-latency ad-hoc querying
  • metrics retention
  • Druid
  • User-space remote PCAP
  • network interactions
  • the fundamental abstraction at which monitoring works
  • adoption of microservices and containerized apps
Listen to your APIs, See your Apps - Netsil
29 annotations
 
blog.netsil.com
611
  • Tracing SDKs and Proxies
  • Application performance management
  • Capacity Planning and Forecasting
  • network or IPC interactions
  • groups of processes or compute instances
  • application maps
  • driving at 100mph with blindfolds!
  • see all the components and their dependencies
  • Application Operations Center
A Comparison of Mapping Approaches for Distributed Cloud Applications
9 annotations
 
blog.netsil.com
602
  • Kubernetes cluster explorer
Announcing New Relic's Kubernetes Cluster Explorer
1 annotations
 
blog.newrelic.com
574
  • 数据服务类
  • SLO 是通过调度来保证的
  • 行存和列存
  • 分析类
  • 复杂查询
  • 点查询
  • 各种查询模式
  • 分析服务
  • 实时数据应用等查询
  • 效果跟踪
  • 实时报表
  • HSAP 系统向上承接了实时大屏
  • 用户、商品、商家属性数据和海量的用户行为数据经过实时和离线的数据清洗统一进入 HSAP 系统
  • 存储计算分离的架构保证了不需要等待移动 / 拷贝数据就能快速完成这些操作
  • 存储计算分离
  • 查询的峰值能达到平均的 3 倍
  • 写入峰值能达到平均的 2.5 倍
  • 弹性和可扩展性的要求
  • HSAP 系统对数据的实时性有着很高的要求
  • 每秒钟数千万甚至上亿条事件
  • 海量数据的实时写入
  • 每秒钟数千万个查询
  • 高并发的混合工作负载
  • HSAP 面临着和已有的大数据以及传统的 OLAP 系统相比很不一样的挑战
  • 具备分析的能力和能够高效地分析并不是一回事
  • 行存的数据格式
  • 实时分析的能力
  • 数据孤岛
  • 另外一个系统提供在线数据服务
  • 业务的洞察
  • 分析海量的数据
  • 服务 / 分析的融合
  • 实时化
  • 批处理
  • 交互式查询
  • 持续聚合
  • 数据服务
  • 预聚合
  • 实时导入
  • 滑动窗口等方式去计算出各种不同维度和时间粒度的特征
  • 大量的用户行为事件
  • 模型的连续更新
  • 实时特征
  • 用户的实时行为做出个性化的推荐
  • HTAP 的架构
  • 数据定期（比如说每天）同步到一个 OLAP 系统
  • 分析型的 OLAP 系统
大数据的下一站是什么？服务/分析一体化（HSAP） - InfoQ
47 annotations
 
www.infoq.cn
653
  • 基础的东西，然后选择你的旅程
  • 文档对用户友好的技
  • 没有知道一切的义务
  • 用jQuery并不蠢
  • 工具是JavaScript最困难的部分之一
  • Babel
  • 构建一切
致程序员：别无谓焦虑，快学习现代 JavaScript 生态体系
7 annotations
 
baijiahao.baidu.com
488
  • If the service name is not registered, the server returns a NOT_FOUND GRPC status
grpc/health-checking.md at v1.15.0 · grpc/grpc
1 annotations
 
github.com
594
  • router rule for the "/" path
  • Well, its the router variable which can call handler functions for specific URLs
  • creates a new connection
  • accepts a request
  • TCP listener
  • requests
  • a listening socket
How Go works with web · Build web application with Golang
7 annotations
 
astaxie.gitbooks.io
586
  • Listener
net - The Go Programming Language
1 annotations
 
golang.org
618
  • connection drops
  • retransmission rates
  • AOC
  • instrumentation of code
  • upfront investment
  • operational metrics
  • Saturation is measured by looking at infrastructure metrics such as Disk I/O, Network I/O and System Memory
  • source-of-truth
  • Latency, Traffic and Errors Tracked Based On Service-Level KPIs
  • granularity
  • Latency, Traffic and Errors Tracked
  • service topology map of the application
  • stream-processing technology
  • service communication protocols
  • run API and services driven cloud applications
  • Netsil Application Operations Center (AOC)
  • Service Level Objectives (SLOs)
  • Services degrade in performance
  • How “full” is the service
  • rate of requests that fail
  • requests per secon
  • The time it takes to service a request
  • saturation
  • errors
  • traffic
  • latency
  • instrumentation of code
  • code-embedded agents
  • golden signals of monitoring that are the foundation of service-level observability for large-scale production applications
The 4 Golden Signals of API Health and Performance in Cloud-Native Applications
29 annotations
 
blog.netsil.com
609
  • Aggregated API Server (AA)
  • Custom Resource Definition (CRD)
  • fit-for-purpose platform
Patterns of Kubernetes API Extensions - ITNEXT
3 annotations
 
itnext.io
635
  • 解决问题
面向TCE的GPU资源池化方案 - Feishu Docs
1 annotations
 
bytedance.feishu.cn
567
  • It provides a way to list, provision, and bind with external Managed ServicesA software offering maintained by a third-party provider. from Service BrokersAn endpoint for a set of Managed Services offered and maintained by a third-party. without needing detailed knowledge about how those services are created or managed
Service Catalog | Kubernetes
1 annotations
 
kubernetes.io
611
  • seconds of ingestion and visualized at the right level of abstraction
  • stream-processing of live service interactions and aggregating analytics results across the environment
  • listening to the application chatter and using that as the “source of truth,”
  • a metrics generator that occasionally streams movies
  • shift to containers and microservices
  • hosts and VMs
  • Framework Diversity
  • Cross-team Collaboration
  • outside the code
  • operational complexity
  • eventual consistency
  • distribution
  • Efficiency
  • Resiliency
  • Freedom of Choice
  • Agility
  • preferred developer choice
  • Operational Challenges
  • microservices
  • monolithic architectures
  • cloud, DevOps and microservices
  • analytics and observability tool
Microservices, DevOps and Production Complexity - Netsil
22 annotations
 
blog.netsil.com
655
  • complicate root-cause analysis
  • a spike in latency on the dependency chain
  • zoom and pan features
  • sock-shop app
  • auto-discovered application maps
  • Application Operations Center
  • application maps
  • absolutely critical to understand the service dependencies in microservices applications
Kubernetes Monitoring Needs Maps - Netsil
8 annotations
 
blog.netsil.com
646
  • https://godoc.byted.org/module-proxy
go module proxy 接入指南 tl;dr 版 - Feishu Docs
1 annotations
 
bytedance.feishu.cn
596
  • request's deadline
  • request-specific values
Go Concurrency Patterns: Context - The Go Blog
2 annotations
 
blog.golang.org
649
  • QoS classes to make decisions about scheduling and evicting Pods
Configure Quality of Service for Pods | Kubernetes
1 annotations
 
kubernetes.io
647
  • GPU池化和虚拟化功能
  • 无法远程调用GPU卡等
  • Pod无法共享G卡
  • 比如仅能使用整数数量的GPU卡资源
  • 同时为这些应用程序提供在GPU资源池中的硬件算力
  • ，
  • 无需挂载物理GPU
  • 高效的通讯机制
  • 几乎零性能损失的情况下，提供对于物理GPU的细粒度、可远程、运行时动态可配置的虚拟化方案
  • GPU资源池化解决方案
  • 混合云+本地数据中心的集中部署与管理
  • 用户提供了更大的价值和更加灵动的任务处理能力，加速推动AI计算与Kubernetes的发展及落地
  • 轻量级GPU虚拟化方案
Rancher与趋动科技达成深度合作，共同打造面向K8S的GPU资源池化方案_RancherLabs_传送门
13 annotations
 
chuansongme.com
640
  • exec 的方式进入一个 pod
  • 资源使用情况
  • 健康状态上面，可以实时地进行观测
从零开始入门 K8s | 可观测性：你的应用健康吗？_Kubernetes中文社区
3 annotations
 
www.kubernetes.org.cn
634
  • k8s的OSB API实现
PaaS平台的服务目录(二): Service Catalog in K8S/OpenShift Origin - 简书
1 annotations
 
www.jianshu.com
642
  • package manager for kubectl plugins
kubernetes-sigs/krew: 📦 Find and install kubectl plugins
1 annotations
 
github.com
584
  • Power tools for kubectl
ahmetb/kubectx: Faster way to switch between clusters and namespaces in kubectl
1 annotations
 
github.com
598
  • The go command resolves imports by using the specific dependency module versions listed in go.mod
  • makes dependency version information explicit and easier to manage
Using Go Modules - The Go Blog
2 annotations
 
blog.golang.org
627
context - The Go Programming Language
0 annotations
 
golang.org
639
  • 全堆栈上工作，并使用端到端方法来处理问题
Hacker Noon创始人访谈:Harmony创始人Stephen Tse_区块链_金色财经
1 annotations
 
www.jinse.com
600
  • memcg
可能是最好的中文cgroup v2文章： Cgroup V2 and writeback support-电子头条-EEWORLD电子工程世界
1 annotations
 
news.eeworld.com.cn
613
  • npm install admin-lte@^3.0 --save
Introduction | AdminLTE 3 Documentation
1 annotations
 
adminlte.io
439
  • /api/v1
  • /apis/metrics.k8s.io/v1beta1
  • Efficient Visualization
  • Occupation of Nodes by Namespaced Pods
  • Daily and Monthly Usage Costs
  • Hourly Usage & Trends
  • Namespace-focused
  • namespace level
  • resource usage reports
  • hourly, daily, monthly
  • capacity planning decisions
  • ease cost allocation
  • long-term usage
  • mid-
  • short-
  • track the resources being consumed by their Kubernetes clusters to prevent overpaying
rchakode/kube-opex-analytics: Kubernetes Cost Allocation and Capacity Planning Analytics Tool. Built-in hourly, daily, monthly reports - Prometheus exporter - Grafana dashboard.
16 annotations
 
github.com
322
  • 给jQuery对象绑定一个新方法是通过扩展$.fn对象实现的
扩展 - 廖雪峰的官方网站
1 annotations
 
www.liaoxuefeng.com
416
  • DOM 操作是 JavaScript 最常见的任务
  • “文档对象模型”（Document Object Model）
DOM 概述 - JavaScript 教程 - 网道
2 annotations
 
wangdoc.com
285
  • For Kubernetes, thinking about the problem as a level triggered system has led to an architecture that is clean, simple
  • Even without disruptions to the network, an edge triggered system trying to reconcile two states could end up with an incorrect outcome
  • reconcile it with the desired state
  • take the actual state
  • the actual state
  • the desired state
  • level triggered view
  • edge triggered view
Level Triggering and Reconciliation in Kubernetes | Hacker Noon
8 annotations
 
hackernoon.com
368
  • Vue.js single file components
  • AdminLTE is the most famous free Bootstrap Admin Template
Creating Adminlte vue components part 0. Introduction.
2 annotations
 
medium.com
420
  • 定义、开发和部署的最小单位
  • 集群里的计算单元是漂浮的
  • 友好的 Web 界面，用于展示和操作 Openshift 的资源
  • ）
  • 镜像仓库（Image Registry）
  • 容器的调度是封装的 Kubernetes
  • Kubernetes概念
  • 底层Docker格式的容器镜像
  • 完整可用的软件系统
  • 数据库服务、日志服务、监控服务
  • 计算、存储、网络以及其他硬件资源
  • 数据中心九层设施
什么是openshift - 菜鸟当家 - 博客园
12 annotations
 
www.cnblogs.com
556
  • consistent hashing
  • responsibility is to place each file on exactly one of its subvolumes
  • Distributed Hash Table
  • Distributed Replicated
  • Distributed
  • Replicated
  • no data redundancy
  • bricks
  • Volume
  • Geo-Replication
  • actual filesystem binary "hello"
  • FUSE module
  • glibc
  • mount point /tmp/fuse
  • API that can be accessed from userspace
  • userspace filesystem
Architecture - Gluster Docs
16 annotations
 
docs.gluster.org
629
  • File System in USErspace
  • don't reside in the kernel
  • The kernel is designed to be stable and provide proper access to resources for user applications
  • move the code to user space
  • delineation
  • User-Space
  • Having file systems in user-space
  • a niche requirement
  • a file system Renaissance
User Space File Systems - Linux Today Blog
9 annotations
 
www.linuxtoday.com
572
  • API that takes the pod descriptor and provides suggested resources requests
  • smart algorithm to calculate memory and cpu values based on historic metrics
  • historic resources usage and OOM events of all pods to suggest new values of the “request” resources spec
  • VPA Recommender
  • VPA can also reacts to OOM (out of memory) events
  • cluster level scalability
  • Pods layer autoscalers
  • Vertical Pod Autoscaler (VPA)
  • Horizontal Pod Autoscaler (HPA)
  • Autoscaling Building Blocks
  • different scalability mechanisms
  • running in a healthy state and running efficiently
  • scale pods and applications
Kubernetes Autoscaling 101: Cluster Autoscaler, Horizontal Pod Autoscaler, and Vertical Pod Autoscaler
13 annotations
 
levelup.gitconnected.com
580
  • 资源的调度，随着数据中心规模进一步扩大，资源的调度会提上日程
  • API可编程来驱动
如何落地全球最大 Kubernetes 生产集群 - 云+社区 - 腾讯云
2 annotations
 
cloud.tencent.com
576
  • Streaming API
  • exec，logs等接口
  • CRI shim就会直接创建出一个轻量级的虚拟机来充当Pod
  • CRI设计的一个重要原则就是确保这个接口本身，只关注容器，不关注POD。
  • SyncLoop本身要求这个控制循环是绝对不可以被阻塞的
  • 2016年，SIG-Node通过将kubelet对容器的操作，统一地抽象成一个接口，具体的接口实现，由各个容器运行时来实现
  • GenericRuntime
  • dockershim
  • CRI shim
Kubernetes(6) SIG-Node与CRI - 简书
9 annotations
 
www.jianshu.com
570
  • provision cloud services directly from the comfort of native Kubernetes tooling
kubernetes-sigs/service-catalog: Consume services in Kubernetes using the Open Service Broker API
1 annotations
 
github.com
604
  • data-card-widget attribute provides cards with the ability to collapse or be removed
  • Cards can contain tools to deploy a specific event or provide simple info
  • Gradient Background Color
  • Background Color
  • Outline
  • contextual classes
  • The card component is the most widely used component through out this template
Cards Component | AdminLTE 3 Documentation
7 annotations
 
adminlte.io
346
  • proportional control
  • Laser
  • Scribe
  • Tupperware
  • pressure stall information (PSI)
  • maximum bandwidth limit.
  • an absolute bandwidth allocation model for realtime scheduling policy
  • normal scheduling policy
  • weight and absolute bandwidth limit models
CPU Controller · cgroup2
9 annotations
 
facebookmicrosites.github.io
625
  • 面向数据查询(ask for exactly what you want)。
(...) RPC vs REST vs GraphQL - 狮乐园 - SegmentFault 思否
1 annotations
 
segmentfault.com
638
  • io.latency
  • You protect workloads with io.latency by specifying a latency target (e.g., 20ms). If the protected workload experiences average completion latency longer than its latency target value, the controller throttles any peers that have a more relaxed latency target than the protected workload. The delta between the prioritized cgroup's target and the targets of other cgroups is used to determine how hard the other cgroups are throttled: If a cgroup with io.latency set to 20ms is prioritized, cgroups with latency targets <= 20ms will never be throttled, while a cgroup with 50ms will get throttled harder than a cgroup with a 30ms target. Interface The interface for io.latency is in a format similar to the other controllers: MAJOR:MINOR target=<target time in microseconds> When io.latency is enabled, you'll see additional stats in io.stat: depth=<integer>—The current queue depth for the group. avg_lat=<time in microseconds>—The running average IO latency for this group. This provides a general idea of the overall latency you can expect for this workload on the specified disk. Note: All cgroup knobs can be configured through systemd. See the systemd.resource-control documentation for details. Using io.latency The limits are applied only at the peer level in the hierarchy. This means that in the diagram below, only groups A, B, and C will influence each other, and groups D and F will influence each other. Group G will influence nobody. Thus, a common way to configure this is to set io.latency in groups A, B, and C. Configuration strategies Generally you don't want to set a value lower than the latency your device supports. Experiment to find the value that works best for your workload: Start at higher than the expected latency for your device, and watch the avg_lat value in io.stat for your workload group to get an idea of the latency during normal operation. Use this value as a basis for your real setting: Try setting it, for example, around 20% higher than the value in io.stat. Experimentation is key here since avg_lat is a running average and subject to statistical anomalies. Setting too tight of a control (i.e., too low of a latency target) provides greater protection to a workload, but it can come at the expense of overall system IO overhead if other workloads get throttled prematurely. Another important factor is that hard disk IO latency can fluctuate greatly: If the latency target is too low, other workloads can get throttled due to normal latency fluctuations, again leading to sub-optimal IO control. Thus, in most cases then, you'll want to set the latency target higher than expected latency to avoid unnecessary throttling—the only question is by how much. Two general approaches have proven most effective: Setting io.latency higher (20-25%) than the usual expected latency. TThis provides a tighter protection guarantee for the workload. However, the tighter control can sometimes mean the system pays more in terms of IO overhead, which leads to lower system-wide IO utilization. A setting like this can be effective for systems with SSDs. Setting io.latency to several times higher than the usual expected latency, especially for hard disks. A hard disk's usual uncontended completion latencies are between 7 and 20ms, but when contention occurs, the completion latency balloons quickly, easily reaching 10 times normal. Because the latency is so volatile, workloads running on hard disks are usually not sensitive to small swings in completion latency; things break down only in extreme conditions when latency jumps several times higher (which isn't difficult to trigger). Effective protection can be achieved in cases like this by setting a relaxed target on the protected group (e.g., 50 or 75ms), and a higher setting for lower priority groups (e.g., an additional 25ms over the higher priority group). This way, the workload can have reasonable protection without significantly compromising hard disk utilization by triggering throttling when it's not necessary. How throttling works io.latency is work conserving: as long as everybody's meeting their latency target, the controller doesn't do anything. Once a group starts missing its target it begins throttling any peer group that has a higher target than itself. This throttling takes two forms: Queue depth throttling—This is the number of outstanding IO's a group is allowed to have. The controller will clamp down relatively quickly, starting at no limit and going all the way down to 1 IO at a time. Artificial delay induction—There are certain types of IO that can't be throttled without possibly affecting higher priority groups adversely. This includes swapping and metadata IO. These types of IO are allowed to occur normally, but they are "charged" to the originating group. Once the victimized group starts meeting its latency target again, it will start unthrottling any peer groups that were throttled previously. If the victimized group simply stops doing IO the global counter will unthrottle appropriately. fbtax2 IO controller configuration As discussed previously, the goal of the fbtax2 cgroup hierarchy was to protect workload.slice. In addition to the memory controller settings, the team found that IO protections were also necessary to make it all work. When memory pressure increases, it often translates into IO pressure. Memory pressure leads to page evictions: the higher the memory pressure, the more page evictions and re-faults, and therefore more IOs. It isn’t hard to generate memory pressure high enough to saturate a disk with IOs, especially the rotating hard disks that were used on the machines in the fbtax2 project. To correct for this, the team used a strategy similar to strategy 2 described above: they prioritized workload.slice by setting its io.latency to higher than expected, to 50ms. This provides more protection for workload.slice than for system.slice, whose io.latency is set to 75ms. When workload.slice has been delayed by lack of IO past its 50ms threshold, it gets IO priority: the kernel limits IO from system.slice and reallocates it to workload.slice so the main workload can keep running. hostcritical.slice was given a similar level of protection as workload.slice since any problems there can also impact the main workload. In this case it used memory.min to guarantee it will have enough to keep running. Though they knew system.slice needed lower IO priority, the team determined the 75ms number through trial and error, modifying it repeatedly until they achieved the right balance between protecting the main workload and ensuring the stability of system.slice. In the final installment of this case study, we'll summarize the strategies used in the fbtax2 project, and look at some of the utilization gains that resulted in Facebook's server farms. ← Memory Controller Strategies and ToolsCPU Controller →cgroup2 IO controller enhancementsInterface filesProtecting workloads with io.latencyInterfaceUsing io.latencyConfiguration strategiesHow throttling works
  • This is where you specify IO limits
  • O
  • accounting of all IOs per-cgroup
  • IOPS
  • system has the flexibility to limit IO to low priority workloads
IO Controller · cgroup2
7 annotations
 
facebookmicrosites.github.io
642
  • some is indicative of added latency due to lack of resources
  • s
  • The full metric indicates the percentage of time in which all tasks were delayed by lack of resources
  • a loss of overall throughput
  • microseconds
Getting Started with PSI · PSI
5 annotations
 
facebookmicrosites.github.io
527
  • 一句话介绍cgroup：把一个cgroup目录中的资源划分给它的子目录，子目录可以把资源继续划分给它的子目录，为子目录分配的资源之和不能超过父目录，进程或者线程可以使用的资源受到它们委身的目录的限制。
  • domain cgroup ，称呼为“ 进程(子)目录 ”
  • threaded cgroup ，称呼为“ 线程子目录
  • cpuacct
  • hugetlb
  • rdma
Linux的资源限制功能cgroups v1和cgroups v2的详细介绍 | 码农网
6 annotations
 
www.codercto.com
594
  • conmon(container monitorring)是一个监控程序,用来作为每个容器的容器引擎（podman ,cri-o）和OCI运行时（runc）之间通信的工具。
  • 我想这是Redhat公司大力推举Podman替换Docker的同时又不是用户体验的诡计
  • building和singing和pushing容器镜像到不同的镜像库中
  • CRI-O的范围与CRI的范围相关
CRI-O vs Podman vs Docker vs CRI-containerd - 简书
4 annotations
 
www.jianshu.com
623
  • with the RPC interface and internal plugins, we get the best of all of these features
  • terraform internal-plugin
  • Statically compiled binaries are much larger than dynamically-linked binarie
  • a network RPC interface instead.
  • statically-compiled binaries
  • Go Plugin Architecture
  • all of the official plugins are shipped as a single binary
  • communicate with the main Terraform binary over an RPC interface
Internal Plugins - Terraform by HashiCorp
8 annotations
 
www.terraform.io
578
  • 云原生时代
  • 支持异构
  • 云原生应用
  • 底层无关
  • 底
  • 硬件配置
  • 云供应商、
  • 调度系统
  • 容器运行时
  • 异构的平台
  • 中间件这种又该怎么描述
  • 描述一个应用
  • 高内聚、低耦合、可扩展、分布式的应用Paas平台
  • 维稳、降本、提效，稳定性应该是运维根本
  • 需求->开发->测试->发布->运维->运营实现全链路的覆盖
  • 应用全生命周期管理
  • 服务化能力
  • 自助化操作
  • 流程平台的联动
从传统Paas到OAM入门篇_Kubernetes中文社区
19 annotations
 
www.kubernetes.org.cn
564
  • 三个大块
  • 一次定义、随处运行
  • Rudr 项目
  • 运维拿到的 API，一定是对 K8s 能力的模块化、声明式的描述
  • 研发拿到的 API 一定是研发视角的
  • API 子集
  • 运维、平台关心的 API 给封装起来
  • 开发者真正关心的 API 给暴露出来
  • K8s API 做“减法”。更准确的说，是我们应该通过“做减法”，把开发者真正关心的 API 给暴露出来，把运维、平台关心的 API 给封装起来
  • 运维能力的描述
  • 上千个字段
  • 定义应用
  • Helm 和 Application CRD 只是单纯地将 K8s 的 API 组合在一起，无法描述我们对云上面资源的依赖
  • Docker 解决了单机应用交付
  • 应用定义”相关的工作
  • 第三层的 OpenKruise
  • 高效、简洁的应用管理与交付体系
  • Operator 以及 K8s 的各种工作负载组件
  • GitOps 的管理
  • 应用交付分层模型
  • K8s 现在并没有一个统一的、平台无关的应用描述能力
  • 插件太多
  • 应用交付场景复杂
  • 设计是给基础设施工程师用来构建其他平台用的（比如 PaaS 或者 Serverless），而不是直面研发和运维同学的
  • Docker
  • 大规模编排调度系统
  • LXC 技术构建容器
给 K8s API “做减法”：阿里巴巴云原生应用管理的挑战和实践
27 annotations
 
mp.weixin.qq.com
605
  • 服务抽象一直是k8s需要增强的特性，然而这也正是CF所提出的OSB API解决的问题
  • k8s在设计之初没有考虑过外部服务接入的问题
  • 应用的打包，依赖管理，迁移以及升级都由Docker解决
  • 主流的PaaS平台那样是面向Container
  • CF在设计之初是面向Application的PaaS
  • Service Broker处于整个系统的服务抽象层
  • OSB API可以将任意软件快速集成到CF环境中
  • 开放服务代理协议
  • 开发者以服务的方式按需消费应用所依赖的基础设施软件
  • PaaS平台以服务的形式来提供
  • 应用托管和迁移的成本
  • service broker所代理的外部服务
PaaS平台的服务目录(一):Service Broker的前世今生 - 简书
12 annotations
 
www.jianshu.com
656
  • package.json文件，并且将入口文件写入main
  • 参数字符串不以“./“或”/“开头，则表示加载的是一个默认提供的核心模块（位于Node的系统安装目录中），或者一个位于各级node_modules目录的已安装模块（全局安装或局部安装）
  • 参数字符串以“./”开头，则表示加载的是一个位于相对路径
  • 参数字符串以“/”开头，则表示加载的是一个位于绝对路径的模块文件
  • 后缀名默认为.js
  • require命令的基本功能是，读入并执行一个JavaScript文件，然后返回该模块的exports对象
  • CommonJS规范加载模块是同步的，也就是说，只有加载完成，才能执行后面的操作
  • 只使用module.exports
  • Node为每个模块提供一个exports变量，指向module.exports
  • module.exports属性表示当前模块对外输出的接口
  • module.exports 表示模块对外输出的值
  • 模块加载的顺序，按照其在代码中出现的顺序
  • 模块可以多次加载，但是只会在第一次加载时运行一次
  • 所有代码都运行在模块作用域，不会污染全局作用域
  • CommonJS规范规定，每个模块内部，module变量代表当前模块。这个变量是一个对象，它的exports属性（即module.exports）是对外的接口。加载某个模块，其实是加载该模块的module.exports属性
CommonJS规范 -- JavaScript 标准参考教程（alpha）
15 annotations
 
javascript.ruanyifeng.com
529
  • maintain continue working well for their users
  • ensure the services they run
  • resource utilization
  • isolate, measure, and control the distribution of resources for a collection of processes on a serve
  • tools and strategies
  • Isolating a core workload from background system resource needs
  • isolate a web server from background system processes
  • separate time-critical or latency-sensitive work from long-term asynchronous jobs
  • pseudo-filesystem cgroupfs
  • cgroup core
  • subsystem controllers
  • hierarchy
  • isolate and protect a system's main workload from widely distributed system binaries and other system services that run on many Facebook hosts
Maximizing Resource Utilization with cgroup2 · cgroup2
13 annotations
 
facebookmicrosites.github.io
596
  • process can be in only one cgroup at a time
  • only the root cgroup exists, to which all processes belong
  • cgroup.controllers
  • cgroup.subtree_control
  • descendant cgroups;
Creating and organizing cgroups · cgroup2
5 annotations
 
facebookmicrosites.github.io
599
  • pressure thresholds
  • in-kernel data structures
  • page cache
  • Shows the total amount of memory currently being
  • system OOM killer will be invoked
  • OOM killer is invoked on the cgroup
  • memory usage hard limit
  • memory.high
Memory Controller · cgroup2
8 annotations
 
facebookmicrosites.github.io
614
  • a memory-intensive process
  • out-of-the-box improvement over the kernel OOM killer
  • The kernel OOM handler’s main job is to protect the kernel
  • oomd
  • rejects a few and continues to run
  • Load shedding
  • oad shedding is a technique to avoid overloading and crashing a system by temporarily rejecting new requests. The idea is that all loads will be better served if the system rejects a few and continues to run, instead of accepting all requests and crashing due to lack of resources. In a recent test, a team at Facebook that runs asynchronous jobs, called Async, used memory pressure as part of a load shedding strategy to reduce the frequency of OOMs. The Async tier runs many short-lived jobs in parallel. Because there was previously no way of knowing how close the system was to invoking the OOM handler, Async hosts experienced excessive OOM kills. Using memory pressure as a proactive indicator of general memory health, Async servers can now estimate, before executing each job, whether the system is likely to have enough memory to run the job to completion. When memory pressure exceeds the specified threshold, the system ignores further requests until conditions stabilize. The chart shows how async responds to changes in memory pressure: when memory.full (in orange) spikes, async sheds jobs back to the async dispatcher, shown by the blue async_execution_decision line. The results were signifcant: Load shedding based on memory pressure decreased memory overflows in the Async tier and increased throughput by 25%. This enabled the Async team to replace larger servers with servers using less memory, while keeping OOMs under control. oomd - memory pressure-based OOM oomd is a new userspace tool similar to the kernel OOM handler, but that uses memory pressure to provide greater control over when processes start getting killed, and which processes are selected. The kernel OOM handler’s main job is to protect the kernel; it’s not concerned with ensuring workload progress or health. Consequently, it’s less than ideal in terms of when and how it operates: It starts killing processes only after failing at multiple attempts to allocate memory, i.e., after a problem is already underway. It selects processes to kill using primitive heuristics, typically killing whichever one frees the most memory. It can fail to start at all when the system is thrashing: memory utilization remains within normal limits, but workloads don't make progress, and the OOM killer never gets invoked to clean up the mess. Lacking knowledge of a process's context or purpose, the OOM killer can even kill vital system processes: When this happens, the system is lost, and the only solution is to reboot, losing whatever was running, and taking tens of minutes to restore the host. Using memory pressure to monitor for memory shortages, oomd can deal more proactively and gracefully with increasing pressure by pausing some tasks to ride out the bump, or by performing a graceful app shutdown with a scheduled restart. In recent tests, oomd was an out-of-the-box improvement over the kernel OOM killer and is now deployed in production on a number of Facebook tiers. Case study: oomd at Facebook See how oomd was deployed in production at Facebook in this case study looking at Facebook's build system, one of the largest services running at Facebook. oomd in the fbtax2 project As discussed previously, the fbtax2 project team prioritized protection of the main workload by using memory.low to soft-guarantee memory to workload.slice, the main workload's cgroup. In this work-conserving model, processes in system.slice could use the memory when the main workload didn't need it. There was a problem though: when a memory-intensive process in system.slice can no longer take memory due to the memory.low protection on workload.slice, the memory contention turns into IO pressure from page faults, which can compromise overall system performance. Because of limits set in system.slice's IO controller (which we'll look at in the next section of this case study) the increased IO pressure causes system.slice to be throttled. The kernel recognizes the slowdown is caused by lack of memory, and memory.pressure rises accordingly. oomd monitors the pressure, and once it exceeds the configured threshold, kills one of the processes—most likely the memory hog in system.slice—and resolves the situation before the excess memory pressure crashes the system. This behavior ← Memory ControllerIO Controller →Memory overcommitPressure-based load sheddingoomd - memory pressure-based OOMCase study: oomd at Facebook
  • outweigh the overhead of occasional OOM events
  • demand exceeds the total memory available
  • Overcommitting on memory—promising more memory for processes than the total system memory—is a key technique for increasing memory utilization
Memory Controller Strategies and Tools · cgroup2
10 annotations
 
facebookmicrosites.github.io
589
  • a transparent and comprehensive resource control framework
  • strategy
fbtax2: Putting it All Together · cgroup2
2 annotations
 
facebookmicrosites.github.io
635
  • MDN 学习区
学习 Web 开发 | MDN
1 annotations
 
developer.mozilla.org
584
大规模场景下 kubernetes 集群的性能优化 - 知乎
0 annotations
 
zhuanlan.zhihu.com
562
  • C4模型没有规定任何特定的符号
  • Code
  • Component diagram
  • Container diagram
  • 人员（角色）和软件系统上
  • 显示系统概况
  • 细节并不重要
  • System Context diagram
  • Context, Container, Component and (optionally) Code(e.g. UML class) diagrams
  • “组件”一词在软件开发行业中是一个非常重载的术语
  • 容器代表托管代码或数据的事物
  • 软件系统是最高级别的抽象
  • （组件）Component
  • 容器）Container
  • 软件系统）Software System
  • （人）Person
  • easy to learn and use
  • reflect how software architects and developers think about and build software
  • abstraction-first
软件架构模型之『C4模型-灵动肆意』 - 知乎
19 annotations
 
zhuanlan.zhihu.com
642
  • walltime
  • 10 seconds, 1 minute, and 5 minutes
  • two metrics: some and full.
  • fair warning of impending resource shortages
  • Pressure Stall Information (PSI)
  • resource pressure
  • memory, CPU, and IO.
  • detect resource shortages
  • pressure stall information
  • accumulated microseconds
PSI Pressure Metrics · cgroup2
10 annotations
 
facebookmicrosites.github.io
635
  • sharing data and results
  • high volume data at low latency
  • stream processing As-a-Service
  • Mantis platform
  • Mantis applications (jobs)
  • an ecosystem of realtime stream processing applications.
Mantis
6 annotations
 
netflix.github.io
501
  • Mantis Runtime
  • resource manager for scheduling Job tasks is Mesos.
Infrastructure Overview - Mantis
2 annotations
 
netflix.github.io
611
  • IO Cache
  • 服务画像的资源分配，对不同服务的容器设定不同的资源限制，除了常见的CPU、内存外，还有IO的限制、ulimit限制、PID限制等等。所以我们扩展了K8s来完成这些工作
  • Cache的数据写到磁盘的速率受容器Cgroup IO配置的限制
  • Buffer IO共享一个内核缓存，并且缓存不隔离，没有速率限制，导致高IO容器很容易影响同主机上的其他容器
  • Direct IO和Buffered IO
美团容器平台架构及容器技术实践_Kubernetes中文社区
5 annotations
 
www.kubernetes.org.cn
614
  • The io controller, in conjunction with the memory controller, implements control of page cache writeback IOs
  • CPU
  • Memory
  • IO
  • PID
  • Cpuset
  • Device
  • RDMA
  • HugeTLB
  • Misc
  • A read-only flat-keyed file
cgroup-v2.rst « admin-guide « Documentation - kernel/git/tj/cgroup.git - cgroup export tree
11 annotations
 
git.kernel.org
642
  • the
enhancements/20191118-cgroups-v2.md at master · kubernetes/enhancements
1 annotations
 
github.com
582
  • 重组知识架构，探求知识背后的本质
  • 扩展性主题阅
  • 理解阅读阶段
  • 框架
  • 扫读：通过标题名称、作者等关键信息快速判断信息内容的类别和主题
  • 不加区分的阅读，很容易陷入知识焦虑
  • 顺手的工具
  • 需要时能及时找到并调用相关知识
  • 隐性知识显性化
  • 碎片知识体系化
  • 外部知识池
  • 大脑是用来思考的，不是用来记忆的
  • 知识隐性化、碎片化
  • 知识管理将是决定竞争力和发展空间的关键
  • 获得、创造、分享、整合、记录、存取、更新、创新
互联网人如何进行知识管理？
15 annotations
 
mbd.baidu.com
632
  • tags are used to designate versions of software within in the repository
  • each layer has tag and a Universally Unique Identifier (UUID)
A Practical Introduction to Container Terminology - Red Hat Developer
2 annotations
 
developers.redhat.com
592
  • kubelet/KubeGenericRuntimeManager
  • RuntimeSpec
  • cri-o
  • CRI-containerd
  • dockershim
  • ImageSpec
扩展 Kubernetes 之 CRI - 云+社区 - 腾讯云
6 annotations
 
cloud.tencent.com
646
  • apiserver
  • StaticPodURL
  • StaticPodPath
深入分析kubelet（1）——获取Pod创建请求 - 简书
3 annotations
 
www.jianshu.com
573
  • 与用户距离之类
  • 网络带宽
  • Filecoin 现在连测试网都没有上，更不可能有用户。只有有用户开始使用 IPFS 服务，上传文件，才能挖矿
  • 把 ASICs 设计成能支持对大量内存的随机访问可能是昂贵且可能是不切实际的
  • 得有用户，有用户上传
  • 检索矿工的最大带宽将决定它的交易总量
  • 检索矿工
  • Proof-of-Work
  • Proof-of-Spacetime
  • Proof-of-Replication
  • Filecoin 就是缺失的激励机制
  • IPFS 是按内容搜索的技术
  • 让 IPFS 协议能够被更多人使用，让更多的人愿意拿出自己的电脑和存储空间给其他人使用
  • 链上流通的 Token 也叫 Filecoin
  • 域名地址
  • 内容地址
  • 相同文件系统的计算机连接起来
  • 全球的、点对点分布式文件存储协议
  • 星际文件系统
  • IPFS 的想法，就是让文件分布式存储和读取
  • IPFS 技术就把文件打碎，分散地存储在不同的硬盘里，下载的时候，再从这些散落在全球各地的硬盘里读取
  • IPFS 其实就是一种 BitTorrent 协议，开发团队对 BitTorrent 协议稍微升级了一下
来看看你对 IPFS 到底误解有多深！
22 annotations
 
baijiahao.baidu.com
612
  • Go和JavaScript
  • 哈希计算
  • 需要数据的时候，通过文件的索引，从原来存储的位置找回来
  • ipfs的工作机制是将整个文件进行拆散，然后储存在全球的不同节点
  • ipfs可以帮助我们节约大量的带宽和存储等
  • 基于BitTorrent的协议
  • FUSE与HTTP
  • 及与内容相关超链接
  • 高吞吐量、按内容寻址的块存储模型
IPFS_全球百科
9 annotations
 
ispeak.vibaike.com
624
  • Support for cgroup v2
opencontainers/runc: CLI tool for spawning and running containers according to the OCI specification
1 annotations
 
github.com
551
kubelet: add initial support for cgroupv2 by giuseppe · Pull Request #85218 · kubernetes/kubernetes
0 annotations
 
github.com
529
  • 内存
  • CPU
  • /proc/[pid]/cgroup
  • systemd 系统对 cgroups 的支持
  • 内存中的临时文件
  • mount | grep cgroup
  • 资源调度控制器(又叫 controllers)
linux cgroups 简介 - 云+社区 - 腾讯云
7 annotations
 
cloud.tencent.com
620
  • CPU/Memory/IO fields are very different from v1
  • containerd metrics interfaces
  • he cgroups lib
cgroups v2 Support · Issue #3726 · containerd/containerd
3 annotations
 
github.com
609
  • 限制cgroup的huge pages的使用量。
  • 限制cgroup访问块设备的IO速度。
  • 用来限制cgroup的CPU使用率。
  • 统计cgroup的CPU的使用率。
  • 绑定cgroup到指定CPUs和NUMA节点。
  • 挂载一棵cgroup树，但不关联任何subsystem，下面就是systemd所用到的方式
  • cgroup virtual filesystem
(...) Linux Cgroup系列（01）：Cgroup概述 - Linux程序员 - SegmentFault 思否
7 annotations
 
segmentfault.com
655
  • runC 是从 Docker 的 libcontainer 中迁移而来的，实现了容器启停、资源隔离等功能
  • 容器配置和标准输入、标准输出、标准错误三个管道文件
  • docker-containerd-shim 另一个参数，是一个和容器相关的目录
  • gRPC 接口，使得 Docker Daemon 屏蔽下面的结构变化
  • 容器执行
  • 镜像管理
  • pstree -l -a -A 20708
  • pstree
Docker、Containerd、RunC...：你应该知道的所有 - InfoQ
8 annotations
 
www.infoq.cn
591
  • OpenShift 作为企业级容器平台，它是一盏打开从 PaaS 到 DevOps 和微服务的大门
  • OCP 从发展来看，是先攻破开发，后打通运维
  • OCP 因 K8S 而重生（OCP3 基于 K8S 重构）、K8S 因 OCP 被企业级客户所接受
  • 企业容器平台就是让客户的应用运行的更快、开发的更快。
  • 敏态、稳态双模 IT 的概念
  • 对于客户而言，如果没有 K8S，那么容器只是“单机版”，很难做到企业级使用
  • 一次打包，到处运行的能力
  • Docker 提供了容器的镜像构建、打包等技术
  • FreeBSD 系统分区为多个子系统
  • 容器时代，大多数是客户自己了解了容器后，寻找能够提供企业级容器解决方案的厂商
  • 容器的起点就是面向应用，而不是像虚拟机那样面向运维
  • 操作系统
  • 容器的应用
  • 操作系统
  • X86 虚拟化是：“虚拟机用假的 CPU、内存和网络欺骗操作系统
  • 操作系统
  • 容器是直接用宿主机的操作系统来欺骗容器的应用”
  • X86 虚拟化技术，不需要改造应用
  • X86 虚拟化是：“虚拟机用假的 CPU、内存和网络欺骗操作系统”
  • OpenShift 的核心价值是什么呢
  • 容器的核心价值是什么呢
  • 企业容器市场如何火热
  • OpenShift 在企业容器市场领导力
  • 2019 年 9 月 IHS Markit Technology 发布了调查报告
  • r2018 年第四季度发布了企业容器平台（ECP）软件套件评估报
  • 2017 年全球企业容器市场
  • 一盏打开从 PaaS 到 DevOps 和微服务的大门
OpenShift 的核心价值到底在哪里？ - InfoQ
27 annotations
 
www.infoq.cn
584
cgroupv2 support meta issue · Issue #2315 · opencontainers/runc
0 annotations
 
github.com
537
  • systemd.unified_cgroup_hierarchy=1
  • kernel 5.3
  • shim v2 when running with cgroup2
  • support cgroup2 (unified mode)
support cgroup2 by AkihiroSuda · Pull Request #40174 · moby/moby
4 annotations
 
github.com
561
  • THP 是一个抽象层，能够自动创建、管理和使用传统大页
  • THP
  • 2MB  和  1GB
  • 4096 bytes 即 4K
Huge pages (标准大页)和 Transparent Huge pages(透明大页)_ITPUB博客
4 annotations
 
blog.itpub.net
626
  • Podman does NOT speak CRI
Crictl Vs Podman
1 annotations
 
www.openshift.com
648
  • the share of resources used by active pods on the node
  • cumulative daily usage
  • cumulative hourly usage
  • hourly
  • capacity planning
  • cost allocation
  • short-, mid- and long-term CPU and memory resource usage analytics
  • cost allocation
  • Usage accounting
Introducing an Analytics Tool for Kubernetes Cost Allocation and Capacity Planning
9 annotations
 
medium.com
370
  • 云计算技术和大数据领域
  • 横向扩展
  • 架构设计、开发方式、部署维护
  • 容器化封装、自动化管理、面向微服务，是云原生系统必须具备的三大特征
  • 敏捷基础设施
  • 微服务
  • 持续交付
  • DevOps
  • 声明式API
  • 不可变基础设施
  • 微服务
  • 服务网格
  • 容器
  • 声明式API
  • 不可变基础设施
  • 微服务
  • 服务网格
  • 容器
  • 等新型动态环境中，构建和运行
  • 公有云、私有云和混合云
  • 可管理性
  • 弹性
  • 扩展性
  • 云原生应用程序凭借其灵活性
  • 应用现代化是绝大多数企业的一项重要工作
  • 数字经济占GDP的比重超过了30%
  • 被动数字化升级
云原生如此重要，可惜80%的人都不知道
27 annotations
 
mbd.baidu.com
617
  • should not typically require special privileges to run their workloads
  •  Container design patterns
  • Storage Driver: devicemapper
  • The graph driver is the piece of software that maps the necessary image layers in the Repository to a piece of local storage
  • optionally a tag
  • a namespace
  • a standard server name
  • docker images
  • a separate copy-on-write layer is also set up
  • aufs, devicemapper, btrfs, zfs, and overlayfs.
  • hostname, etc
  • network interfaces
  • user ids
  • process ids
  • virtual mount points
  • clone()
  • each “commit” creates a new layer
  • A namespace is a tool for separating groups of repositories
  • each directive in the file creates a new layer
  • Runc is a very terse utility and expects a mount point (directory) and meta-data (config.json) to be provided to it
  • runc
  • libcontainer
  • LXC
  • Setting up cgroups
  • container metadata
  • container mount point
  • isolated
  • runtime instantiation of a Container Image
  • a lower level component
  • docker, RKT, CRI-O, and LXD
  • OCI-based images
  • architecture of the building blocks of containers
  • Container Image formats
  • Container Engines
  • Open Containers or the OCI
  • The process for starting containers, as well as the image format on disk, are defined and governed by standards.
  • running
  • rest
  • conceptual differences
A Practical Introduction to Container Terminology - Red Hat Developer
39 annotations
 
developers.redhat.com
597
  • runC其实就是在libcontainer的基础上进行了封装成各个Command
runC源码分析——主体调用链 - 云+社区 - 腾讯云
1 annotations
 
cloud.tencent.com
558
  • podWorkers
  • NewMainKubelet
  • CreateAndInitKubelet
  • RunKubelet
  • 该接口用于从kubelet命令行获取参数
  • KubeletConfiguration
(...) Kubelet源码分析(一):启动流程分析 - kubernetes - SegmentFault 思否
6 annotations
 
segmentfault.com
521
  • distpatchWork 创建 pod
  • NewMainKubelet
  • 节点机的Root容器，采集并分析该节点机的全面的使用情况
  • 自动采集CPU、内存、文件系统和网络使用的统计信息
  • 分析容器资源使用率和性能特性的代理工具
  • ReadinessProbe
  • LivenessProbe
  • 容器计算一个hash值
  • 创建一个Pause容器
  • Secret
  • 挂载外部卷
  • 数据目录
  • WatchAndList
Kubernetes 源码分析之 Kubelet - 简书
13 annotations
 
www.jianshu.com
563
  • unpack that image into an OCI Runtime filesystem bundle
  • OCI Image
  • filesystem bundle
  • the Image Specification
  • the Runtime Specification
About the Open Container Initiative - Open Container Initiative
5 annotations
 
opencontainers.org
656
  • A Picture is Worth a Thousand Words
  • fine-grained context
  • Scaling Challenge
  • ad hoc dynamic queries on live event streams
  • Event Stream Processing
  • Operational insight
  • detect, diagnose, and resolve problems
  • cohesive set of tools
  • alerting mechanisms
  • dashboards
  • proactively detect
  • exceptional
  • high availability across such a complicated system
  • large, complex, and highly distributed
Improving Netflix’s Operational Visibility with Real-Time Insight Tools
14 annotations
 
netflixtechblog.com
591
  • Preserving operational visibility and response capabilities
  • memorialization
  • uniqueness
  • post-incident analysis
  • Technical Sleuthing
  • Technical Sleut
  • Scribe
  • remediation
  • Decision Making
  • Coordination
  • Incident management
  • Incident ma
  • Incident Manager
  • monitor high-level business KPIs
  • they operate what they build
  • reliability consulting
  • handling the lifecycle of an incident
  • systemic risk identification
  • ensuring Netflix stays up
  • business-critical operations
  • robust and reliable
  • reliability
  • best-in-class service
Keeping Customers Streaming — The Centralized Site Reliability Practice at Netflix
23 annotations
 
netflixtechblog.com
591
  • Monte Carlo matching procedure
  • geographic dimensions
  • biased treatment populations
  • Together
  • Through being proactive rather than reacting to issues
  • Through being proactive rather than reacting to issues
  • best possible experience
  • understand the effects of in-field software changes on performance metrics
  • Comparing metrics between two web browser software versions or operating system versions
  • 0.5%
  • a change
  • treatment samples,
  • reatment samples
  • treatment sample
  • matched control
  • metric distributions
  • data sparsity issues
  • a matched control sample
  • Monte Carlo
  • Jigsaw
  • methodology
  • treatment sample distribution
  • geographic region
  • a random subsample of the control sample
  • geographic region
  • population of the same devices
  • control sample
  • Detecting Performance Anomalies
  • change to the app or backend servers
  • network behavior interact
  • device characteristics
  • ecosystem continues to grow both in volume and diversity
Detecting Performance Anomalies in External Firmware Deployments
32 annotations
 
netflixtechblog.com
614
  • 大量随机样本，去了解一个系统
蒙特卡罗方法入门 - 阮一峰的网络日志
1 annotations
 
www.ruanyifeng.com
582
  • 均值
  • 3个标准差
  • 标准差
  • 均值
  • 蒙特卡洛模拟有个更为直观清晰的认识
  • 通过大量成千上万甚至百万次的模拟次数，最终得出一个累计概率分布图，这个就是蒙特卡洛模拟
  • 蒙特卡洛模拟正是以概率为基础的一种方法，所以用赌城的名字为这种方法命名
蒙特卡洛模拟(Monte Carlo Simulation)浅析 - 知乎
7 annotations
 
zhuanlan.zhihu.com
599
Study
0 annotations
 
study.bytedance.net
645
  • kuberuntime_manager.go
  • 创建cgroups
  • syncPod
深入分析kubelet（2）——创建Pod - 简书
3 annotations
 
www.jianshu.com
639
  • cgroup v2 (unified mode)
  • are not considered to be production
runc/cgroup-v2.md at master · opencontainers/runc
2 annotations
 
github.com
640
  • 随机数（或更常见的伪随机数）来解决一些复杂的计算问题
采样之蒙特卡罗方法 - 小糊涂也学要编程 - 博客园
1 annotations
 
www.cnblogs.com
604
  • Mann-Whitney U test
Automated Canary Analysis at Netflix with Kayenta - Netflix TechBlog
1 annotations
 
netflixtechblog.com
615
  • nonparametric statistical methods
  • fast-paced system experiments
  • daily or weekly patterns
  • account for
  • randomized A/B experiments
  • understand the impact of QoE
  • deduction-induction iterations
  • deduction and induction
  • System Experiments
  • modeling projects
  • modeling aspect
  • statistical models
  • statistical challenges
  • continuous experimentation
  • data-driven decision
  • data-driven decisio
  • devices
  • network infrastructure
  • varying taste
  • quality of experience (QoE)
A/B Testing and Beyond: Improving the Netflix Streaming Experience with Experimentation and Data Science
20 annotations
 
netflixtechblog.com
639
  • DBSCAN
  • Density-Based Spatial Clustering of Applications with Noise (DBSCAN)
  • unsupervised machine learning technique
  • use cluster analysis
  • improving reliability
  • not bad enough to be detected by a threshold aler
  • automate the detection of servers that are operating in a degraded state
  • periodic tuning
  • A very unhealthy server
  • A slow or unhealthy server is worse than a down server
  • Is there a way to use time series data to automatically find these outliers
  • elevated request processing latency
  • less than one percent of those become unhealthy
  • automated outlier detection
  • look beyond
  • go beyond
  • heightened senses
  • one rogue server
Tracking down the Villains: Outlier Detection at Netflix
18 annotations
 
netflixtechblog.com
564
  • custom pattern detection algorithms
  • which page
  • Priam side car
  • rich operational events
  • intuitive UI that allows SREs
  • matching a certain criterion
  • explore events
  • operators
  • anomalies
  • millions of interactions
  • Double Exponential Smoothing (DES)
  • mantis-publish library
  • abnormal change
  • non-comprehensive
Use cases - Mantis
14 annotations
 
netflix.github.io
600
  • stream-processing evolve as an important and complementary tool in our operational and insights toolset at Netflix
  • insights capabilities
  • time to recover, or TTR
  • time to detect (TTD)
  • key performance indicator
  • rolling windows of unique events per asset
  • fine-grained insights
  • anomalies in high-volume, high-cardinality data in realtime
  • alerting on individual video titles across countries and device
  • API enables easy programmatic integration with automated workflows
  • The UI gives users the ability to quickly and manually interact with jobs and platform functionality
  • API for Integration
  • UI for Self-service
  • individual job’s autoscaling in action
  • autoscales the Mesos worker cluster
  • Mantis autoscales both the cluster size and the individual jobs as needed
  • peak capacity all the time
  • anomaly detector application
  • efficient data and code reuse
  • error rates per device type
  • stateful computations
  • sink to collect and output the processed data
  • processing stages
  • source
  • high-cardinality event streams
  • high-volume
  • a flexible model for defining a stream-processing job
  • a shared pool of computing resources
  • cloud native
  • long tail of problems
  • granularity
  • too little at the same time
  • too much data
  • high throughput
  • low-latency
  • speed, power, and incredible visibility
  • Mantis shrimp
  • a stream-processing service platform
  • 8 million events per second
  • insights
  • better visibility into our complex operational environments
Stream-processing with Mantis - Netflix TechBlog
41 annotations
 
netflixtechblog.com
611
  • Vue 实例还提供了一些有用的实例属性与方法。它们都有前缀 $，以便与用户定义的属性区分开来
  • 个 Vue 实例被创建时，它向 Vue 的响应式系统中加入了其 data 对象中能找到的所有的属性。当这些属性的值发生改变时，html 视图将也会产生相应的变化
Vue.js 起步 | 菜鸟教程
2 annotations
 
www.runoob.com
400
  • evolve separately without any dependency
  • a common approach
  • data format
  • formats
  • naming conventions
  • APIs interface for resources inside the system
  • 6 architectural constraints
  • Representational State Transfer
  • high level design guidelines
REST Principles and Architectural Constraints – REST API Tutorial
9 annotations
 
restfulapi.net
512
  • CRI + containerd ShimV2 revolution
  • Container Runtime management engine
  • Sigma/Kubernetes
  • lower-layer Container Runtime
  • CRI + containerd shimv2
  • CRI is the first calling interface in Kubernetes to be divided into plug-ins
  • remove and decouple the complex features that are originally invasive to the main code from the core library one by one by dividing them into different interfaces and plug-ins
  • how to connect containerd to the kata container
  • implementation of Shimv2 API
  • kata-Containerd-Shimv2
  • container-shim-v2 in Sandbox
  • a containerd shim
  • specify a shim for each Pod
  • containerd shim for each container
  • make KataContainers follow containerd
  • standard interface between the CRI shim and the containerd runtime
  • Containerd ShimV2
  • CRI-O
  • reuse the existing CRI shims
  • What can a CRI shim do? It can translate CRI requests into Runtime APIs
  • CRI shim
  • Dockershim
  • maintenance
  • we do not want a project like Docker to have to know what a Pod is and expose the API of a Pod
  • Containerd-centric API
  • Container Runtime Interface
  • multi-tenant
  • security
  • Kernel version run by your container is completely different from that run by the Host machine
  • each Pod now has an Independent kernel
  • the more layers you build here, the worse your container performance is
  • SECCOMP
  • secure Container Runtime
  • we are concerned about security
  • each Pod like the KataContainer is a lightweight virtual machine with a complete Linux kernel
  • a compressed package of your program + data + all dependencies + all directory files
  • the Container Image
  • the Container Runtime
  • runC that helps you set up these namespaces and cgroups, and helps you chroot, building a container required by an application
  • binding operation
  • NodeName field of the Pod object
  • Pods are created, instead of containers
  • the designs of Kubernetes CRI and Containerd ShimV2
  • KataContainers
  • RuntimeClass
  • ShimV2
  • container runtime
  • CRI
  • design and implementation of key technical features
CRI and ShimV2: A New Idea for Kubernetes Integrating Container Runtime - Alibaba Cloud Community
49 annotations
 
www.alibabacloud.com
493
  • eliminates an extra hop in the stack,
  • A container runtime is software that executes containers and manages container images on a node
  • cri-containerd,
  • execute containers
  • manages images
  • Image Management
  • an implementation of CRI for containerd
Containerd Brings More Container Runtime Options for Kubernetes | Kubernetes
7 annotations
 
kubernetes.io
548
  • NodeLease
  • at scale for thousands of clusters today
  • blast-radius of potential hick-ups
  • gardener-scheduler
  • custom controller-manager that acts on our own custom resources
  • operator pattern
Gardenlet by rfranzke · Pull Request #1601 · gardener/gardener
6 annotations
 
github.com
628
  • alias docker=podman
  • daemonless, open source, Linux native tool
What is Podman? — Podman documentation
2 annotations
 
docs.podman.io
512
  • https://github.com/opencontainers/runtime-spec
containers/crun: A fast and lightweight fully featured OCI runtime and C library for running containers
1 annotations
 
github.com
581
  • host-specific details
  • containing low-level OS
  • container configuration
  • OCI-compliant bundles
  • OCI configuration file
  • Application Bundle Builders
opencontainers/runtime-spec: OCI Runtime Specification
6 annotations
 
github.com
505
  • operational insight applications
  • ad-hoc exploration
  • tens of minutes to a few seconds
  • Mean-Time-To-Detect
  • realtime nature
  • rich and relevant context
  • anomalies
  • trigger and alert relevant teams within seconds
  • early warning indicator
  • identifies negative impact on the viewing experience with fine-grained granularity
  • several critical operational insight applications
  • as much context
  • derive new insights with this data
  • rigor
  • ahead of time
  • without having to add new instrumentation to your applications
  • ask new questions
  • one at a time
  • access these events in realtime
  • data in its original form is already lost
  • We should have access to raw events
  • minimize the costs of observing and operating
  • detrimental
  • streaming-based observability on events
  • better understand the behavior
  • on-demand
  • Cost-Effective
  • tap into an event stream from any instrumented application
  • democratization
  • The traditional way of working with metrics and logs alone is not sufficient for large-scale and growing systems
  • minimize the costs of observing and operating complex distributed systems
  • streaming microservices ecosystem
  • Platform
  • Platform
Open Sourcing Mantis: A Platform For Building Cost-Effective, Realtime, Operations-Focused Applications
34 annotations
 
netflixtechblog.com
515
  • shim grpc service is the main source of interaction with the shim
  • Outputs
  • Inputs
  • shim should not write to any other location on disk except the bundle
  • OCI bundle is still the main source of configuration for shims
  • state and abstract actions
  • various runtimes
  • The goals is to provide an API that various runtimes can implement to add support in containerd while still having control of state and abstract actions
  • a shim API
  • state management
  • More VM based runtimes have internal state and more abstract actions
[proposal] Shim API v2 · Issue #2426 · containerd/containerd
11 annotations
 
github.com
519
  • execution lifecycle of a container
  • a first class shim API for runtime authors to integrate with containerd
Runtime v2 (shim API) by crosbymichael · Pull Request #2434 · containerd/containerd
2 annotations
 
github.com
536
  • Lazy显存分配方式
  • GPU卡多个任务间的显存隔离
  • 限制任务的最大显存策略
  • 一张GPU卡可以同时运行多个深度学习训练/预测任务
  • 按需自动分配显存
  • 显存自增长AutoGrowth策略
  • 有限的显存支持多个开发者同时进行实验，执行各自的任务
  • 一张卡或一台机器上同时承载更多的训练和预测任务
  • 高效地利用GPU显存
  • 深度学习模型的训练和推理往往需要大量的显存，用来支持更大的模型和更大的batch size
穷！深度学习中如何更好地利用显存资源？
10 annotations
 
mp.weixin.qq.com
259
  • Alternate Logo
Main Sidebar Component | AdminLTE 3 Documentation
1 annotations
 
adminlte.io
411
  • animation
  • tool tips
  • every aspect of your charts
  • light-weight
  • flexible
  • responsive
  • HTML5 canvas based
Fancy, Responsive Charts with Chart.js — SitePoint
7 annotations
 
www.sitepoint.com
236
  • drain the Node from its workloads, remove it from the cluster and re-join it.
  • systemd as the cgroup driver stabilized the system
  • two views of those resources
  • two different cgroup managers
  • cgroupfs
  • Systemd has a tight integration with cgroups and will allocate cgroups per process
  • Cgroup drivers
Container runtimes | Kubernetes
7 annotations
 
kubernetes.io
385
  • competencies
  • ML models
  • data products
  • data exploration
  • ETL
  • symbiotic partnership
  • onus
  • constructing derived metrics
  • creating baselines
  • segments
  • slicing the data
  • the last decade
  • 1990s
  • concepts and methods to improve business decision making by using fact-based support systems
Fire Your BI Team - Towards Data Science
14 annotations
 
towardsdatascience.com
267
  • clean and easy way
  • a decorator pattern
  • batching
  • go’s concurrency model
  • transforming
  • moving
  • how to build data pipelines using golang
  • data driven applications
Concurrent data pipelines in golang | Towards Data Science
8 annotations
 
towardsdatascience.com
647
  • libcontainer
  • execdriver作为Docker容器的执行驱动，负责创建容器运行命名空间，负责容器资源使用的统计与限制，负责容器内部进程的真正运行等
  • execdriver
  • networkdriver
  • graphdriver
  • Driver是Docker架构中的驱动模块。通过Driver驱动，Docker可以实现对Docker容器执行环境的定制
终于有人把 Docker 讲清楚了，万字详解
6 annotations
 
www.toutiao.com
394
  • ResourceQuota
  • medium
  • huge pages do not support overcommit
  • discover and report all huge page resources as schedulable resources
  • pre-allocate huge pages for multiple sizes
  • huge page capacity
  • pre-allocate huge pages
Manage HugePages | Kubernetes
7 annotations
 
kubernetes.io
359
  • Error Boundary 标签
  • vue提供的handleError句柄
  • sourceMap就是一个文件，里面储存着位置信息。
  • sourceMap
  • 前端工程化的时代,工程化导出的代码一般都是被压缩混淆后的
  • Webpack工程化
  • 返回true 就不会被上抛了
  • window.onerror
从0到1，Vue大牛的前端搭建——异常监控系统 - 知乎
8 annotations
 
zhuanlan.zhihu.com
223
  • i
  • no conversion was needed for pids and hugepages, correct?
  • more to do beside adding support for hugetlb once it lands in a kernel release
  • Support for running on a host that uses cgroups v2 unified mode
  • conversion from the cgroups v1 limits to cgroups v2
kubelet: add initial support for cgroupv2 by giuseppe · Pull Request #85218 · kubernetes/kubernetes
5 annotations
 
github.com
393
  • 抓手、生态、闭环、拉齐、梳理、迭代、owner意识
  • 能说、会写、善做是对职场人的三大要求
  • 撕逼甩锅邀功抢活这些闹心的事儿基本也不会缺席
  • PPT、沟通、表达、时间管理、设计、文档等方面的能力
  • 报警配置和监控梳理
  • 良好的规划能力和清晰的演进蓝图
  • 做系统建设要有全局视野
  • 有的人能把一个小盘子越做越大
  • 想到了leader没想到的地方
  • 直接去找对应的人聊，让别人讲一遍自己基本就全懂了，这效率比看文档看代码快多了
  • 向上沟通反馈
  • owner意识
  • 主动承担任务，主动沟通交流，主动推动项目进展，主动协调资源，主动向上反馈，主动创造影响力
  • 主动承担，及时交流反馈
  • 主动跳出自己的舒适区，感到挣扎与压力的时候，往往是黎明前的黑暗，那才是成长最快的时候
  • 强迫自己跳出自己的安逸区
  • 积极学习，保持技术能力、知识储备与工作年限成正比，这到了35岁哪还有什么焦虑呢
  • 架构先行于业务
  • 技术同学该如何培养产品思维，引导产品走向
  • 系统建设？系统核心能力，系统边界，系统瓶颈，服务分层拆分，服务治理
  • 代码层，可以做的事情更多了，资源池化、对象复用、无锁化设计、大key拆分、延迟处理、编码压缩、gc调优还有各种语言相关的高性能实践
  • 在架构层，可以做缓存、预处理、读写分离、异步、并行等等
  • 术到道的过程
  • 知识还是零星的几点，不成体系，不仅容易遗忘，而且造成自己视野比较窄，看问题比较局限
在滴滴和字节跳动干了 2 年后端开发，太真实… - InfoQ 写作平台
24 annotations
 
xie.infoq.cn
386
  • Resource Limitation
  • Prioritization
  • Control
  • Accounting
  • 进程（task）在 fork 自身时创建的子任务（child task）默认与原 task 在同一个 cgroup 中，但是 child task 允许被移动到不同的 cgroup 中
  • 这个 cgroup 也称为 root cgroup
  • 系统每次新建一个 hierarchy 时，该系统上的所有 task 默认构成了这个新建的 hierarchy 的初始化 cgroup
  • 同一个 hierarchy 可以附加一个或多个 subsystem
  • Resource Controller
  • hierarchy（层级树）
  • subsystem（子系统
  • cgroup（控制组）
  • task（任务）
  • cgroups 可以限制、记录、隔离进程组所使用的物理资源（包括：CPU、memory、IO 等），为容器实现虚拟化提供了基本保证，是构建 Docker 等一系列虚拟化管理工具的基石
  • 进程放到一个组里面统一加以控制
Docker背后的内核知识——cgroups资源限制 - InfoQ
15 annotations
 
www.infoq.cn
337
  • 每个 group 都会包含一个只读文件 cgroup.stat
  • Cgroups v2 只能将进程绑定到叶子节点。因此不能绑定进程到任何一个已开启 controller 的任何 subgroup 中
  • read-only 文件
  • cgroup.subtree_control
  • cgroup.controllers
  • 一个 contoller 无法在 Cgroups v1 和 v2 中同时使用
  • 当 cgroup 为空时的通知机制得到改进，通过 cgroup.events 文件通知
  • v1 版本中的 task 文件和 cpuset controller 中的 cgroup.clone_children 文件被移除
  • 通过 cgroup.controllers 和 cgroup.subtree_control 指定哪些 controller 可以被使用
  • Proess 只能绑定到 cgroup 的根 (“/“) 目录和 cgroup 目录树中的叶子节点
  • Cgroups v2 中所有的 controller 都会被挂载到一个 unified hierarchy 下
  • 允许 Cgroups v1 和 v2 两个版本共存
  • Cgroups v2 将多 hierarchy 的方式变成了 unified hierarchy，并将所有的 controller 挂载到一个 unified hierarchy
  • 实际使用中，通常是每个 hierarchy 一个控制器
  • 控制器 (controller) 只能属于一个 hierarchy
  • 相比 v1 具体有哪些变化等方面
浅谈Cgroups V2 - InfoQ
16 annotations
 
www.infoq.cn
370
  • changes through the entire stack
  • any configuration file designed for cgroups v1 will continue to work on cgroups v2
  • Privileged pods will still use the host cgroup namespace so to have visibility on all the other cgroups
  • hugetlb
  • same naming
  • different naming
  • CPU pinning
  • CPU stats
  • cAdvisor already supports cgroups v2
  • runc support for cgroups v2
  • crun
  • dockershim implementation embedded in the Kubelet
  • support features only available in cgroups v2 but not in cgroups v1, the OCI runtime specs must be changed
  • Device plugins that require v2 enablement
  • OCI runtime specifications
  • run on a host using either cgroups v1 or v2
  • features parity between cgroup v2 and v1
  • Plugins support
  • Dockershim
  • Expose new cgroup2-only features
  • Add support for cgroups v2 to the Kubelet
enhancements/20191118-cgroups-v2.md at master · kubernetes/enhancements
21 annotations
 
github.com
402
  • Podman 不再使用守护进程管理服务
  • 传输过来的打包文件对容器进行恢复。
  • 打包到指定位置
  • 设置检查点
  • Podman 支持将容器从一台机器迁移到另一台机器。
  • Podman CLI 里面87%的指令都和DOcker CLI 相同
  • podman 不需要
  • docker需要使用root用户来创建容器，但是podman不需要
  • podman不支持，如果在k8s中就不存在这个问题，我们可以设置pod的重启策略，在系统中我们可以采用编写systemd服务来完成自启动
  • Podman 可以管理和运行任何符合 OCI（Open Container Initiative）规范的容器和容器镜像
podman初试-和docker对比-首富手记-51CTO博客
10 annotations
 
blog.51cto.com
334
flink.hyrax.tce_trace_sink - Grafana
0 annotations
 
grafana.bytedance.net
390
  • #85218
  • huge page sizes changes
kubernetes/CHANGELOG-1.19.md at master · kubernetes/kubernetes
2 annotations
 
github.com
426
  • 读取100列中的5列，这将帮助你最少减少20倍的I/O消耗。
  • 分析
  • 每秒数百个查询或更少
  • 宽表
  • 少量的列
  • 大量的行
  • 不修改已添加的数据
  • 相当大的批(> 1000 rows)进行写入
  • 大多数是读请求
  • 列式数据库总是将同一列的数据存储在一起，不同列的数据也总是分开存储
  • 联机分析(OLAP)的列式数据库管理系统(DBMS)
什么是ClickHouse？ | ClickHouse文档
11 annotations
 
clickhouse.tech
366
  • 子单元通过 prop 接口与父单元进行了良好的解耦
  • todo 对象是变量，即其内容可以是动态的。 我们也需要为每个组件提供一个“key”，稍后再 作详细解释。
  • 还可以绑定到 DOM 结构
  • 不仅可以把数据绑定到 DOM 文本或 attribute
  • v-bind attribute 被称为指令
  • Vue.component
介绍 — Vue.js
6 annotations
 
cn.vuejs.org
447
  • 插件是有点笨重，但终于可以使用OneNote保存代码注释应该是相当开心的一件事。OneNote是学生记笔记的最佳方式之一，对于计算机和工科学生来说，这是一个特别好的消息。
  • OneNote可以说是最好的笔记应用程序，但目前它缺乏语法高亮功能，因此不适合程序员使用。幸运的是，由GitHub用户发布的免费插件使OneNote可以在代码中突出语法了。
编程时，你都用什么软件做的笔记？还是说你都不做笔记？ - 简书
2 annotations
 
www.jianshu.com
449
  • checkpoint timeout:
Apache Flink 1.10 Documentation: Checkpointing
1 annotations
 
ci.apache.org
416
  • each operator gets the complete list of state elements
  • Union redistribution
  • Even-split redistribution
  • a List of serializable objects
  • list-style managed operator state is supported
  • different redistribution schemes
  • CheckpointedFunction
  • each operator state is bound to one parallel operator instance
  • Keyed State is always relative to keys and can only be used in functions and operators on a KeyedStrea
Apache Flink 1.10 Documentation: Working with State
9 annotations
 
ci.apache.org
472
  • The container runtime can start a separate streaming server upon request (and can potentially account the resource usage to the pod!), and return the location of the server to kubelet. Kubelet then returns this information to the Kubernetes API server, which opens a streaming connection directly to the runtime-provided server and connects it to the client.
  • it can become a bottleneck for the network traffic on the node
  • kubelet handles the connection of all streaming requests
  • Using tools on the node is not a portable solution because most tools assume the pod is isolated using Linux namespaces
  • level triggered
  • Pod-level features
  • a wider spectrum of runtimes
  • Kubelet is responsible for managing the lifecycles of the containers through the RPCs, exercising the container lifecycles hooks and liveness/readiness checks, while adhering to the restart policy of the pod
  • pod-level cgroup that kubelet creates and passes to the runtime
  • PodSandbox
  • A Pod is composed of a group of application containers in an isolated environment with resource constraints
  • --image-service-endpoint
  • --container-runtime-endpoint
  • interact with containers (exec/attach/port-forward)
  • manage the lifecycle of the pods and containers
  • RuntimeService
  • ImageService
  • CRI shim
  • pluggable container runtimes
  • clearly-defined abstraction layer
  • CRI consists of a protocol buffers and gRPC API, and libraries, with additional specifications and tools under active development
Introducing Container Runtime Interface (CRI) in Kubernetes | Kubernetes
21 annotations
 
kubernetes.io
447
Container Runtimes Part 1: An Introduction to Container Runtimes - Ian Lewis
0 annotations
 
www.ianlewis.org
439
  • runc is currently the most widely used container runtime
  • runc
Container Runtimes Part 2: Anatomy of a Low-Level Container Runtime - Ian Lewis
2 annotations
 
www.ianlewis.org
443
  • OCI runtime bundle
  • containerd is a high-level runtime that was split off from Docker
  • docker-containerd and docker-runc are just Docker packaged versions of vanilla containerd and runc
  • containerd
  • runc
  • building, packaging, sharing, and running containers
  • Docker
Container Runtimes Part 3: High-Level Runtimes - Ian Lewis
7 annotations
 
www.ianlewis.org
521
  • Containerd
  • CRI-O
  • Docker
  • Cgroup drivers
Container runtimes | Kubernetes
4 annotations
 
kubernetes.io
472
  • Docker itself isn't necessary to support CRI
  • a shim between the kubelet and Docker
  • runtime handler is exposed in the Kubernetes API using the RuntimeClass object
  • shim to start the container
  • runtime handler
  • configure crictl to connect to containerd
  • cri-o
  • Docker
  • containerd
  • CRI is simply an abstraction layer or API that allows you to switch out container runtime implementations instead of having them built into the kubelet
  • CRI and introduce a few of the runtimes that support CRI
  • a bridge between the kubelet and the container runtime
Container Runtimes Part 4: Kubernetes Container Runtimes & CRI - Ian Lewis
12 annotations
 
www.ianlewis.org
416
  • taskmanager.memory.segment-size
  • network stack and the memory manager
Apache Flink 1.10 Documentation: Configuration
2 annotations
 
ci.apache.org
408
  • crictl is a command-line interface for CRI-compatible container runtimes
Debugging Kubernetes nodes with crictl | Kubernetes
1 annotations
 
kubernetes.io
477
  • Loading Style
  • display statistical snippets
  • small boxes
  • info boxes
Boxes Components | AdminLTE 3 Documentation
4 annotations
 
adminlte.io
300
  • 多篇并行的过程
  • 记录所找的文章和阅读进度的工具，做好编录工作
  • 保持节奏稳定的持续学习比短期集中精力的高强度更重要
  • 阅读整篇文章
  • 论文主体部分
  • 图片
  • 结论
  • 介绍
  • 作者对于一张可视化图片或者表格花的精力有时甚至超过全篇文字叙述，论文中的图片无疑会成为我们理解文章的钥匙
  • 大致的理解
  • 读标题，摘要，图片
(5 封私信 / 29 条消息) 科研大牛们怎么读文献？ - 知乎
11 annotations
 
www.zhihu.com
526
  • Package flag implements command-line flag parsing
flag - The Go Programming Language
1 annotations
 
golang.org
463
  • --manifest-url-header
kubelet | Kubernetes
1 annotations
 
kubernetes.io
518
  • vue init webpack my-project
  • common workflow recipes
Introduction · GitBook
2 annotations
 
vuejs-templates.github.io
441
  • .navbar-* with .navbar-light or .navbar-dark.
  • layout-boxed
Layout | AdminLTE 3 Documentation
2 annotations
 
adminlte.io
469
  • CORS是一个W3C标准，全称是“跨域资源共享”（跨源资源共享）
  • 同源是指，域名，协议，端口均为相同
  • 面对各种端
  • 微服务概
  • 数据中心化
  • nginx配置
  • 再继续发送对应的GET/POST/PUT/DELETE请求
  • 204或200状态码
  • 非简单请求
  • OPTIONS请求
  • HEAD、GET、POST三种方法之一
  • 禁止使用XHR对象向不同源的服务器发起http请求
  • 跨域资源共享
前后端分离项目，如何解决跨域问题？-java架构设计的回答-悟空问答
13 annotations
 
www.wukong.com
412
  • overall runtime of batch jobs
Predictive CPU isolation of containers at Netflix - Netflix TechBlog
1 annotations
 
netflixtechblog.com
511
  • 传统监控概念的智慧升级
  • 从各种维度观察服务运行的状态是否良好
  • 指标日志、调用链、变更数据等不同角度运维数据在内的统一
  • 仅仅依靠单纯的性能监控指标远远不够
  • 体量规模增大，业务细分的颗粒度也随之提升
  • 云原生微服务架构
  • 调用链基础上的拓扑与图谱推导都会被华丽丽的实现
  • 非监督学习的算法
  • 日志分析
  • 监控指标
  • 运用机器学习的方法来提升运维效率
  • 人工智能技术与基础运维能力的完美集合
想不到！智能运维的正确姿势：从临场救火到淡然饮茶 - 知乎
12 annotations
 
zhuanlan.zhihu.com
483
  • vue里面使用sass的话就更简单了
  • vue create testvue
  • npm install -g @vue/cli
  • npm是node的包管理工具
  • 变量，嵌套，循环，继承，混合
  • css预处理器
如何五分钟搞定Vue + Sass？ - 知乎
6 annotations
 
zhuanlan.zhihu.com
473
  • 可观测数据的微服务依赖关系评分规则
  • 手工治理方法
  • 结合流程和业务团队进行梳理盘点
  • 马科夫链
ServiceRank: 使用马科夫链对微服务进行影响力评分 - 知乎
4 annotations
 
zhuanlan.zhihu.com
567
  • npx 让我放弃了使用 yarn run 这个命令
  • 软链接(windows 快捷方式)的方式组合到 node_modules 中
  • 扁平化安装到一个位置
  • npm 的安装策略并不是扁平化的
  • 语义化的版本号安装应用
  • yarn.lock 文件来管理依赖版本问题，保证每次安装都是一致的
  • npm 来接管全局模块管理
(...) 为什么我从 npm 到 yarn 再到 npm? - 个人文章 - SegmentFault 思否
7 annotations
 
segmentfault.com
479
  • brew uninstall --ignore-dependencies node icu4c
  • node
  • brew install node
dyld: Library not loaded: /usr/local/opt/icu4c/lib/libicui18n.60.dylib
3 annotations
 
gist.github.com
544
  • <router-view>
  • 注册了一个router
  • 改造main.js
  • 不同的页面其实是一个组件，而不是通过去后台请求发来的HTML/FTL文件
[SpringBoot+VueJs] 3.3 VueJs的一些基础知识 - 简书
4 annotations
 
www.jianshu.com
506
  • To specify ranges you need to pass and array where each element is an array with exactly two Date objects (from, to) or their timestamp equivalent.
Date range picker | Vue2 date range picker
1 annotations
 
innologica.github.io
523
  • v-model 指令在表单 <input>、<textarea> 及 <select> 元素上创建双向数据绑定
表单输入绑定 — Vue.js
1 annotations
 
cn.vuejs.org
461
  • 钩子函数
  • 在实例使用 directives 选项来注册局部指令
  • 注册一个全局指令 v-focu
Vue.js 自定义指令 | 菜鸟教程
3 annotations
 
www.runoob.com
627
  • $0 是当前我们选中的 html 节点的引用
【译】你不知道的Chrome调试工具技巧 第一天：console中的 ' - 掘金 - 掘金
1 annotations
 
juejin.im
351
  • 图表的颜色和边框宽度
[转]Chart.js入门教程 - h2z - 博客园
1 annotations
 
www.cnblogs.com
353
  • 唯一 key attribute
  • 你也可以用 v-for 来遍历一个对象的 property。
  • 在 v-for 块中，我们可以访问所有父作用域的 property
列表渲染 — Vue.js
3 annotations
 
cn.vuejs.org
311
  • 写入不少于1000行的批量写入
  • 每秒处理数百个查询
  • 吞吐量可以使用每秒处理的行数或每秒处理的字节数来衡量
性能 | ClickHouse文档
3 annotations
 
clickhouse.tech
337
  • Naming conventions
  • StatusServiceUnavailable
  • lower case and be valid DNS subdomain
  • CamelCase and singular
  • all lowercase and plural
  • a subdomain your group or organization owns, such as "widget.mycompany.com
  • domain name form - the Kubernetes project
  • "pod status" resource (that acts on "Pod" kind)
  • the kind "Pod" is exposed as a "pods" resource
  • API Group
  • Resource
  • Kind
  • should
  • schema
  • those APIs preferentially accept and return JSON
  • clients create, update, delete, or retrieve a description of an object via the standard HTTP verbs (POST, PUT, DELETE, and GET)
  • ease client development
  • The general style of the Kubernetes API is RESTful - clients create, update, delete, or retrieve a description of an object via the standard HTTP verbs (POST, PUT, DELETE, and GET) - and those APIs preferentially accept and return JSON. Kubernetes also exposes additional endpoints for non-standard verbs and allows alternative content types. All of the JSON accepted and returned by the server has a schema, identified by the "kind" and "apiVersion" fields. Where relevant HTTP header fields exist, they should mirror the content of JSON fields, but the information should not be represented only in the HTTP header.
community/api-conventions.md at master · kubernetes/community
18 annotations
 
github.com
371
  • architectural style
  • Uniform Resource Identifiers (URIs)
  • Till the time, you are honoring the 6 guiding principles of REST, you can call your interface RESTful
  • Cacheable
  • REpresentational State Transfer
  • distributed hypermedia systems
  • Roy Fielding
  • portability
  • scalability
  • contain all of the information necessary to understand the request
  • Uniform interface
  • identification of resources
  • manipulation of resources through representations
  • self-descriptive messages
  • hypermedia as the engine of application state
  • REST is defined by four interface constraints
  • constraining component behavior
  • Stateless
  • Client–server
  • Layered system
  • Code on demand
  • The key abstraction of information in REST is a resource
  • resource identifier
  • components
  • resource representation
  • A representation consists of data, metadata
  • hypermedia links
  • media type
  • simultaneous presentation of information and controls
  • self-descriptive
  • one media-type associated with one resource
  • Every media type defines a default processing mode
  • HTML
  • Resource
  • Resource Methods
  • uniform interface
  • all application state transitions must be driven by client selection of server-provided choices that are present in the received representations
  • implied by the user’s manipulation of those representations
What is REST – Learn to create timeless REST APIs
38 annotations
 
restfulapi.net
358
  • 周同比
  • 日环比
  • 物化视图和聚合表引擎，按照指定的时间粒度和维度指标预聚合我们的查询指标，从而降低查询的压力
  • 多副本的聚合 MergeTree 引擎
  • 预聚合的方式来计算一个接口的平均耗时
  • 物化视图功能
  • 数据抽样
  • 分钟粒度
  • 秒级粒度
微博基于ClickHouse灵活监控百亿流量下的业务指标 - 云+社区 - 腾讯云
9 annotations
 
cloud.tencent.com
383
  • col-md-offset-
  • 偏移列
Bootstrap 网格系统 | 菜鸟教程
2 annotations
 
www.runoob.com
336
  • configureWebpack
  • 整 we
webpack 相关 | Vue CLI
2 annotations
 
cli.vuejs.org
513
  • css-loader 和 style-loader
  • webpack 根据模块的依赖关系进行静态分析
Webpack 入门教程 | 菜鸟教程
2 annotations
 
www.runoob.com
361
  • 转译器(transpiler)来支持未来的语言特性
  • m
  • task runner
  • 模块打包器
  • 包管理器
  • task runner(npm脚本)
  • babel或TypeScript
  • 转译代码意味着将一种语言的代码转换为另一种语言的代码
  • 模块打包工具
  • HTML定义了一个叫 moment的全局变量
  • JavaScript最初被设计成只能在浏览器中运行，出于安全的考虑，它不能访问客户端计算机的文件系统，因此JavaScript不支持这一特性
  • moment.min.js在index.js加载之前被加载
【译】向恐龙解释现代JavaScript - 知乎
12 annotations
 
zhuanlan.zhihu.com
320
  • ES6标准引入了rest参数
  • arguments类似Array但它不是一个Array
  • 永远指向当前函数的调用者传入的所有参数
  • 只在函数内部起作用
  • 免费赠送的关键字arguments
函数定义和调用 - 廖雪峰的官方网站
5 annotations
 
www.liaoxuefeng.com
268
  • 为什么在 HTML 中监听事件
  • 事件修饰符
  • 事件处理方法
  • 可以用 v-on 指令监听 DOM 事件
事件处理 — Vue.js
4 annotations
 
cn.vuejs.org
427
  • a general reusable solution
Design Patterns for Humans
1 annotations
 
roadmap.sh
267
  • 机器人交互的体验是否贴心，意图识别是否准确，最终还是要看在 NLP 这块
  • 取决于底层对接的运维平台是否足够全
  • 对用户屏蔽了一系列复杂的运维平台概 念
  • 机器人服务向下打通了十几个底层基础设施平台，同时提供了统一的交互方式给到真实用 户，解决了长期以来运维难、应急响应慢的痛点
  • 统一的 Controller 进行消息接收
  • 消息收发
  • 监控预警
  • VIP 查询
  • 域名查询
  • 系统网络
  • 应用查询
  • 变更操作
  • 运维操作
  • 反馈给
  • 运维对象
  • 智能运维机器人
  • ChatBot 作为应用场景之一
  • 日志查询分析
  • 变更查询
  • 业务链路
  • 监控预警
  • 1 分钟发现、5 分钟定位、10 分钟恢复
详解 Ops 智能运维机器人，故障处理又快又准！
22 annotations
 
mbd.baidu.com
282
  • quota太小导致整机死锁问题
  • period/quota控制需要遍历整个cgroup树影响性能问题
  • 在线无法及时抢占离线的问题
  • 对时延不敏感的业务上，有一定效果
  • cgroup提供的share以及period/quota功能来实现
  • 并没有真正的达到混部的效果，在在线的那些核上，还是没有办法混部离线业务
  • 在线业务限定在某些核上，离线业务限定在某些核上面
  • Cpuset方案
  • 在不影响在线业务的前提下，对整机CPU利用率提升效果非常明显，在有的业务场景下，整机CPU利用率甚至能提升至90%
  • 整机的CPU利用率翻一翻
腾讯成本优化黑科技：整机CPU利用率最高提升至90% - 云+社区 - 腾讯云
10 annotations
 
cloud.tencent.com
320
  • 平均1小时
  • 内存过度分配的数量与CPU相当
  • 统计复用来过度分配资源
  • 尽力而为批处理层的消耗增加
  • 每个小时的平均计算和内存使用情况
Next Borg：谷歌新一代Borg系统
5 annotations
 
mp.weixin.qq.com
310
  • 修饰符 (modifier)
  • 单个表达式
  • JavaScript 表达式
  • v-bind 指令
  • v-html
  • “Mustache”语法 (双大括号) 的文本插值
模板语法 — Vue.js
6 annotations
 
cn.vuejs.org
379
  • 侦听器
  • 侦听属性
  • 缓存
  • 计算属性
计算属性和侦听器 — Vue.js
4 annotations
 
cn.vuejs.org
414
  • v-bind 用于 class 和 style 时，Vue.js 做了专门的增强。表达式结果的类型除了字符串之外，还可以是对象或数组
Class 与 Style 绑定 — Vue.js
1 annotations
 
cn.vuejs.org
420
  • 用 key 管理可复用的元素
条件渲染 — Vue.js
1 annotations
 
cn.vuejs.org
422
  • line chart allows a number of properties to be specified for each dataset
Line · Chart.js documentation
1 annotations
 
www.chartjs.org
420
  • 无酸味、以苦味为主，苦味会加重
  • 　　咖啡的味道更浓，酸味变淡，这是一般咖啡豆的烘焙方法
  • 一般通俗的煎培程度、留有强烈的酸味
  • 烘焙程度越浅，酸味就越浓
  • 烘焙程度越深苦味越浓
  • 生豆占60%，烘焙占30%，萃取占10%
  • 　　所谓咖啡烘焙（coffee roasting）是指通过对生豆的加热，促使咖啡豆内外部发生一系列物理和化学反应，并在此过程中生成咖啡的酸，苦，甘等多种味道，形成醇度和色调，将生豆转化为深褐色原豆的过程
咖啡豆中度烘焙和深度烘焙的区别 深度烘焙和轻度烘焙哪个更好？ 中国咖啡网 gafei.com
7 annotations
 
www.gafei.com
244
  • 但总的来说，对于大多数方法来说，您可能不希望强度超过1:12，因为结果太混乱了。低于1:20左右时，您会得到水质，平淡的咖啡，可能太苦了
  • 对于大多数酿造方法和口味，您可以从1:15到1:18的比例开始。第一个应该给您浓浓的风味咖啡。第二个会更轻
咖啡最佳冲煮比例 咖啡粉水比 滴漏咖啡 gafei.com
2 annotations
 
m.gafei.com
340
  • 本地部署环境即服务
  • 公有云端的部署
  • 分析、数据湖和数据仓库
  • 数据分析工具
  • 先进的架构平台
  • 没有分析和提炼的数据毫无价值
  • 云-边-端的解决方案将驱动基础架构的创新
  • 数据量和数据流的新变化正在重塑IT基础架构
  • 数据增长呈井喷态势
  • 数字经济时代，数据已成为最新和最重要的生产要素
  • 大数据、云计算、物联网、边缘计算、人工智能等领域的技术突破与发展，工业大数据与这些新技术领域的联系将更加紧密，使工业大数据的数据获取量更大，存储管理更便捷，分析产出更智能，由此实现最大化的商业价值
  • 20%的业务将利用它实现实时结果
  • 到2024年，由人工智能自动化、物联网和智能设备需求驱动的数据量将超过30ZB
“新基建”重新定义未来大数据基础架构
13 annotations
 
mbd.baidu.com
324
  • export default命令用于指定模块的默认输出。显然，一个模块只能有一个默认输出，因此export default命令只能使用一次
  • export default关键字对外导出一个匿名函数，导入这个模块的时候，可以为这个匿名函数取任意的名字
  • 默认导出(export default)
  • import语句会执行所加载的模块
  • import命令具有提升效果，会提升到整个模块的头部，首先执行
  • from指定模块文件的位置，可以是相对路径，也可以是绝对路径，.js路径可以省略
  • import命令接受一对大括号，里面指定要从其他模块导入的变量名
  • import命令也是如此
  • export命令可以出现在模块的任何位置，只要处于模块顶层就可以
  • 通常情况下，export输出的变量就是本来的名字，但是可以使用as关键字重命名
  • 命名空间
  • CommonJS
  • CommonJS 需要一个兼容的脚本加载器作为前提条件
  • require 和 module.exports 的函数
  • ES6模块(module)
  • ES6 的模块自动采用严格模式
  • 作为一个模块，它可以选择性地给其他模块暴露（提供）自己的属性和方法
很全很全的 JavaScript 模块讲解 - 简书
17 annotations
 
www.jianshu.com
286
  • console对象是 JavaScript 的原生对象
console 对象与控制台 - JavaScript 教程 - 网道
1 annotations
 
wangdoc.com
330
  • V8 (Chrome, Chromium)
  • 现代浏览器改为采用“即时编译”（Just In Time compiler，缩写 JIT），即字节码只在运行时编译，用到哪一行就编译哪一行，并且把编译结果缓存（inline cache）
  • 浏览器都将 JavaScript 进行一定程度的编译，生成类似字节码（bytecode）的中间代码，以提高运行速度
  • JavaScript 是一种解释型语言，也就是说，它不需要编译，由解释器实时运行
  • 渲染树绘制到屏幕
  • 渲染树的布局（layout）
  • 将 DOM 和 CSSOM 合成一棵渲染树
  • CSS 代码解析为 CSSOM
  • HTML 代码解析为 DOM
  • Chrome：Blink 引擎
  • 解析和执行 CSS，也会产生阻塞
  • DOMContentLoaded事件
  • 较好的做法是将<script>标签都放在页面底部，而不是头部
  • 如果外部脚本加载时间很长（一直无法完成下载），那么浏览器就会一直等待脚本下载完成，造成网页长时间失去响应，浏览器就会呈现“假死”状态，这被称为“阻塞效应”
  • 加载外部脚本时，浏览器会暂停页面渲染，等待脚本下载并执行完成后，再继续渲染
  • 为了防止攻击者篡改外部脚本，script标签允许设置一个integrity属性，写入该外部脚本的 Hash 签名，用来验证脚本的一致性
浏览器环境概述 - JavaScript 教程 - 网道
16 annotations
 
wangdoc.com
252
  • 一个变量如果未声明，那么默认就是顶层对象的属性
  • 浏览器里面，window对象（注意，w为小写）指当前的浏览器窗口
window 对象 - JavaScript 教程 - 网道
2 annotations
 
wangdoc.com
287
  • 定
Vue.js 教程
1 annotations
 
learning.dcloud.io
237
  • b-row b-col
  • bootstrap-vue bootstrap axios
  • Bootstrap是最受欢迎的独立 CSS 框架
(...) 快速上手BootstrapVue - 疯狂的技术宅 - SegmentFault 思否
3 annotations
 
segmentfault.com
356
  • 想对路由参数的变化作出响应的话，你可以简单地 watch (监测变化) $route 对象
  • 不过，这也意味着组件的生命周期钩子不会再被调用
  • 动态路径参数”(dynamic segment)
动态路由匹配 | Vue Router
3 annotations
 
router.vuejs.org
247
  • Require JS
  • Common JS
  • Script Tag
Integration · Chart.js documentation
3 annotations
 
www.chartjs.org
456
  • 动态组件
  • 插槽分发内容
  • v-model
  • 单个根元素
  • Prop 向子组件传递数据
  • 一个组件的 data 选项必须是一个函数
组件基础 — Vue.js
6 annotations
 
cn.vuejs.org
475
  • 一个组件的 data 选项必须是一个函数，因此每个实例可以维护一份被返回对象的独立的拷贝
组件基础 — Vue.js
1 annotations
 
cn.vuejs.org
492
  • 监控及运维管控更加实时精准
  • 内核隔离更加精细
  • 调度精度更细致
  • 调度层面时效更实时
  • 业务的资源画像刻画得更加细致
数据中心日均 CPU 利用率 45% 的运行之道--阿里巴巴规模化混部技术演进 - IT程序猿
5 annotations
 
www.itcodemonkey.com
420
  • 如何从VS Code跳转到node_modules中的文件定义| 经验摘录qa.1r1g.com › askqa.1r1g.com › askCached
vscode css node_modules 跳转 - Google Search
1 annotations
 
www.google.com
461
  • main header contains the navbar
Main Header Component | AdminLTE 3 Documentation
1 annotations
 
adminlte.io
405
  • Command+Shift+P 调用命令板
vscode使用经验 | 十年一刻
1 annotations
 
hellogithub2014.github.io
435
Control Sidebar Component | AdminLTE 3 Documentation
0 annotations
 
adminlte.io
458
  • 局部安装
  • CLI 服务是构建于 webpack 和 webpack-dev-server 之上的
  • 零配置原型开发
  • 开发环境依赖
  • 全局安装的 npm 包
  • vue create
介绍 | Vue CLI
6 annotations
 
cli.vuejs.org
515
  • public/index.html
HTML 和静态资源 | Vue CLI
1 annotations
 
cli.vuejs.org
448
  • scss/ and js/ are the source code for our CSS and JavaScript
Contents · Bootstrap v4.5
1 annotations
 
getbootstrap.com
426
  • 可以根据本地的文件结构用相对路径来引用静态资源
CSS 相关 | Vue CLI
1 annotations
 
cli.vuejs.org
406
  • Less Loader Sass Loader Stylus Loader
  • Sass 和 Less 之类的预处理器是对原生CSS的拓展
  • style-loader将所有的计算后的样式加入页面中
  • css-loader使你能够使用类似@import 和 url(...)的方法实现 require()的功能
  • css-loader 和 style-loader
  • 模块化，让我们可以把复杂的程序细化为小的文件
  • 开发的文件往往需要进行额外的处理才能让浏览器识别,而手动处理又是非常繁琐的，这就为WebPack类的工具的出现提供了需求
  • 模块打包机：它做的事情是，分析你的项目结构，找到JavaScript模块以及其它的一些浏览器不能直接运行的拓展语言（Scss，TypeScript等），并将其转换和打包为合适的格式供浏览器使用
  • 指定入口文件后，webpack将自动识别项目所依赖的其它文件
  • Babel其实是一个编译JavaScript的平台
  • Babel其实是几个模块化的包，其核心功能位于称为babel-core的npm包中
  • 解析JSX
  • ES6以及JSX
  • 对于模块，Webpack能提供非常强大的处理功能
  • Webpack有一个不可不说的优点，它把所有的文件都都当做模块处理
  • 产品阶段，可能还需要对打包的文件进行额外的处理，比如说优化，压缩，缓存以及分离CSS和JS
(...) 入门 Webpack，看这篇就够了 - 前端学习笔记 - SegmentFault 思否
16 annotations
 
segmentfault.com
440
  • ，loader 类似于其他构建工具中“任务(task)”
  • loader 甚至允许你直接在 JavaScript 模块中 import CSS文件
  • loader 可以使你在 import 或"加载"模块时预处理文件
loader | webpack 中文网
3 annotations
 
www.webpackjs.com
553
  • a variety of ways to organize code into modules
  • special libraries to load modules on demand
  • AMD – one of the most ancient module systems, initially implemented by the library require.js.
  • CommonJS – the module system created for Node.js server.
  • UMD – one more module system, suggested as a universal one, compatible with AMD and CommonJS.
  • language-level module system
  • A module is just a file. One script is one module.
  • ules
Modules, introduction
8 annotations
 
javascript.info
423
  • template:'app/src/page/index.html'
  • filename:'app/index.html',
  • html-webpack-plugin
webpack入门之打包html,css,js,img(一) - 甜玉米莱莱 - 博客园
3 annotations
 
www.cnblogs.com
424
  • extract-text-webpack-plugin
  • style-loader css-loader  
webpack入门之打包html,css,js,img(二) - 甜玉米莱莱 - 博客园
2 annotations
 
www.cnblogs.com
403
  • 构建过程中，资源链接会被自动注入
HTML 和静态资源 | Vue CLI
1 annotations
 
cli.vuejs.org
437
  • 任何放置在 public 文件夹的静态资源都会被简单的复制，而不经过 webpack
  • 调整 webpack 配置最简单的方式就是在 vue.config.js 中的 configureWebpack 选项提供一个对象：
  • webpack
HTML 和静态资源 | Vue CLI
3 annotations
 
cli.vuejs.org
453
  • Loaders和Plugins常常被弄混，但是他们其实是完全不同的东西，可以这么来说，loaders是在打包构建过程中用来处理源文件的（JSX，Scss，Less..），一次处理一个，插件并不直接操作单个文件，它直接对整个构建过程其作用
  • 插件（Plugins）是用来拓展Webpack功能的
  • ，css会和js打包到同一个文件中，并不会打包为一个单独的css文件，不过通过合适的配置webpack也可以把css打包为单独的文件的
  • import, require, url
  • import './main.css';//使用require导入css文件
  • css-loader 和 style-loader
  • eval-source-map
  • Source Maps
  • webpack的配置文件中配置source maps，需要配置devtool
(...) 入门 Webpack，看这篇就够了 - 前端学习笔记 - SegmentFault 思否
9 annotations
 
segmentfault.com
464
  • {Object | Function} plugin
  • 安装 Vue.js 插件
  • Vue.use( plugin )
API — Vue.js
3 annotations
 
cn.vuejs.org
446
  • 例如，url(./image.png) 会被翻译为 require('./image.png')
HTML 和静态资源 | Vue CLI
1 annotations
 
cli.vuejs.org
490
  • npm内置了两个命令简写，npm test等同于执行npm run test，npm start等同于执行npm run start
  • npm run-script lint或者npm run lint
  • npm run-script lint或者npm run lin
  • npm run-script lint或者npm run lin
  • package.json文件有一个scripts字段
  • package.json文件有一个scripts字
  • npm不仅可以用于模块管理，还可以用于执行脚本。packa
npm模块管理器 -- JavaScript 标准参考教程（alpha）
7 annotations
 
javascript.ruanyifeng.com
508
  • 插件(plugins)
  • loader
  • 输出(output)
  • 入口(entry)
  • webpack v4.0.0 开始，可以不用引入一个配置文件。然而，webpack 仍然还是高度可配置的
概念 | webpack 中文网
5 annotations
 
www.webpackjs.com
471
  • JSX
(...) 入门 Webpack，看这篇就够了 - 前端学习笔记 - SegmentFault 思否
1 annotations
 
segmentfault.com
527
  • npm (node package manager) is the dependency/package manager
  • npx your-package
  • npx is also a CLI tool whose purpose is to make it easy to install and manage dependencies hosted in the npm registry
npm vs npx — What’s the Difference?
3 annotations
 
www.freecodecamp.org
475
  • 命令行工具 (CLI)
  • 注册表（registry）
  • 网站
  • 世界上最大的软件注册表
01 - npm 是什么？ | npm 中文文档
4 annotations
 
www.npmjs.cn
357
  • npm install npm@latest -g.
  • installing a Node Version Manager or NVM
  • install npm via the Node.js website
02 - 如何安装 npm 并管理 npm 版本 | npm 中文文档
3 annotations
 
www.npmjs.cn
449
  • 语义化版本
13 - 如何使用语义化版本 | npm 中文文档
1 annotations
 
www.npmjs.cn
442
  • Publisher: IEEE Cite This Cite This PDF 5 Author(s) Arun Kejariwal ; Winston Lee ; Owen Vallis ; Jordan Hochenbaum ; Bryce Yan All Authors Sign In or Purchase to View Full Text 267 Full Text Views Email Export to Collabratec Alerts Alerts Manage Content Alerts Add to Citation Alerts Abstract Document Sections I. Introduction II. Chiffchaff: Design and Use III. Previous Work IV. Conclusion Authors Figures References Keywords Metrics More Like This Footnotes Download PDF Download Citation View References Email Request Permissions Export to Collabratec Alerts Abstract: `Anywhere, Anytime and Any Device' is often used to characterize the next generation Internet. Achieving the above in light of the increasing use of the Internet worldwid... View more Metadata Abstract: `Anywhere, Anytime and Any Device' is often used to characterize the next generation Internet. Achieving the above in light of the increasing use of the Internet worldwide, especially fueled by mobile Internet usage, and the exponential growth in the number of connected devices is non-trivial. In particular, the three As require development of infrastructure which is highly available, performant and scalable. Additionally, from a corporate standpoint, high efficiency is of utmost importance. To facilitate high availability, deep observability of physical, system and application metrics and analytics support, say for systematic capacity planning, is needed. Although there exist many commercial services to assist observability in the data center, public/private cloud, they lack analytics support. To this end, we developed a framework at Twitter, called Chiffchaff, to drive capacity planning in light of a growing user base. Specifically, the framework provides support for automatic mining of application metrics and subsequent visualization of trends (for example, Week-over-Week (WoW), Month-overMonth (MoM)), data distribution etcetera. Further, the framework enables deep diving into traffic patterns, which can be used to guide load balancing in shared systems. We illustrate the use of Chiffchaff with production traffic. Published in: 2013 IEEE 16th International Conference on Computational Science and Engineering Date of Conference: 3-5 Dec. 2013 Date Added to IEEE Xplore: 06 March 2014 Electronic ISBN: 978-0-7695-5096-1 INSPEC Accession Number: 14145932 DOI: 10.1109/CSE.2013.133 Publisher: IEEE Conference Location: Sydney, NSW, Australia googletag.cmd.push(function() { googletag.display("div-gpt-ad-1527780343424-0"); }); googletag.cmd.push(function() { googletag.display("div-gpt-ad-1527780523694-0"); }); Contents I. IntroductionRecent years have seen robust growth - 8% YoY - in Internet usage worldwide [1]. The proliferation of mobile devices such as smartphones and tablets has further fueled Internet usage. Although this lends itself as a huge opportunity for application developers and monetization via advertising et cetera (recently, Berg Insight forecasted that the direct App store revenue would grow at a compound annual growth rate of 40.7 % to reach C=8.8{\rm C}\!\!\!\!\!{=}8.8 billion by 2015 (from C=1.6{\rm C}\!\!\!\!\!{=}1.6 billion in 2010)), it exerts tremendous pressure on the infrastructure services. To this end, in [2], Barroso and Hölzle argue: Sign in to Continue Reading Authors Figures References Keywords Metrics Footnotes googletag.cmd.push(function() { googletag.display("div-gpt-ad-1531324580046-0"); }); More Like This Delay-Aware Resource Allocation for Data Analysis in Cloud-Edge System 2018 IEEE Intl Conf on Parallel & Distributed Processing with Applications, Ubiquitous Computing & Communications, Big Data & Cloud Computing, Social Computing & Networking, Sustainable Computing & Communications (ISPA/IUCC/BDCloud/SocialCom/SustainCom) Published: 2018 Multi-objective Optimization Based Virtual Resource Allocation Strategy for Cloud Computing 2012 IEEE/ACIS 11th International Conference on Computer and Information Science Published: 2012 Show More Top Organizations with Patents on Technologies Mentioned in This Article googletag.cmd.push(function() { googletag.display("div-gpt-ad-1531324238723-0"); });
  • Week-over-Week (WoW
  • Month-overMonth (MoM
Visual Analytics Framework for Cloud Infrastructure Data - IEEE Conference Publication
3 annotations
 
ieeexplore.ieee.org
419
  • Most npm packages are modules
  • module is any file or directory that can be loaded by Node.js' require()
  • package is a file or directory that is described by a package.json
19 - Understanding packages and modules | npm 中文文档
3 annotations
 
www.npmjs.cn
440
JSX_百度百科
0 annotations
 
baike.baidu.com
451
BootstrapVue-基于VueJS的Bootstrap框架 · 全球最流行的 HTML、CSS 和 JS 工具库。
0 annotations
 
code.z01.com
394
  • require('bootstrap') will load all of Bootstrap’s jQuery plugins onto the jQuery object
  • JavaScript plugins
Download · Bootstrap v4.5
2 annotations
 
getbootstrap.com
456
  • bootstrap.min.js
  • popper.min.js
  • jquery-3.5.1.slim.min.js
  • bootstrap.min.css
Bootstrap · The most popular HTML, CSS, and JS library in the world.
4 annotations
 
getbootstrap.com
425
  • examples/ of Bootstrap usage
  • site/docs/ folder includes the source code for our documentation
  • dist/ folder includes everything listed in the precompiled download section above
  • scss/ and js/ are the source code for our CSS and JavaScript
  • jQuery
  • Popper
  • bootstrap.bundle.js bootstrap.bundle.min.js
  • bootstrap.css bootstrap.min.css
  • Utilities
  • Components
  • Content
  • Layout
Contents · Bootstrap v4.5
12 annotations
 
getbootstrap.com
511
  • /js/dist/*.js
  • JavaScript plugins built on jQuery
JavaScript · Bootstrap v4.5
2 annotations
 
getbootstrap.com
502
  • use your existing rule for css without any special modifications to webpack config
  • Importing Compiled CSS
  • npm install --save jquery popper.js
  • import plugins individually
  • Install bootstrap as a Node.js module using npm
  • Import Bootstrap’s JavaScript
Webpack · Bootstrap v4.5
6 annotations
 
getbootstrap.com
481
  • downloadable package contains both of these libraries
Dependencies & Plugins | AdminLTE 3 Documentation
1 annotations
 
adminlte.io
366
  • npm install admin-lte@^3.0 --save
ColorlibHQ/AdminLTE: AdminLTE - Free admin dashboard template based on Bootstrap 4
1 annotations
 
github.com
430
  • <router-view></router-view>
  • a new temple view
  • new route for the new sidebar item
  • bootstrap.js
  • app.js
  • app.scss
  • vue-router
  • a vue-router implementation on sidebar with dynamic content section
Laravel 6 — VueJS — AdminLTE 3 - CrydeTaan - Medium
8 annotations
 
medium.com
494
  • Extra large
  • Small
  • Large
  • Medium
  • .container-{breakpoint},
  • .container-fluid
  • .container
  • Containers are the most basic layout element in Bootstrap
Overview · Bootstrap v4.5
8 annotations
 
getbootstrap.com
388
  • flexbox
  • spacing
  • aligning
  • hiding,
  • showing
Utilities for layout · Bootstrap v4.5
5 annotations
 
getbootstrap.com
410
  • display属性设置为"none"，或把visibility属性设置为"hidden"
CSS Display(显示) 与 Visibility（可见性） | 菜鸟教程
1 annotations
 
www.runoob.com
460
  • 样式
  • 组件
  • 布局
Bootstrap 精华一页纸 - 知乎
3 annotations
 
zhuanlan.zhihu.com
409
  • @vue/cli
安装 | Vue CLI
1 annotations
 
cli.vuejs.org
498
  • 你可以使用 vue serve 和 vue build 命令对单个 *.vue 文件进行快速原型开发
快速原型开发 | Vue CLI
1 annotations
 
cli.vuejs.org
424
  • preset 将会存在用户的 home 目录下一个名为 .vuerc 的 JSON 文件
创建一个项目 | Vue CLI
1 annotations
 
cli.vuejs.org
497
  • 整个MVVM整合Admin-LTE成功
  • 挂载vue.app
  • 改造页面，使之能够挂载我们的vue app
[SpringBoot+VueJs] 3.2 Admin-LTE介绍 - 简书
3 annotations
 
www.jianshu.com
485
  • Linux的VFS很值得欣赏吗？Page Cache在NVMEM上一点用没有，inode里面放那么多namespace这样的数据结构就是作死
“优秀架构设计” - 知乎
1 annotations
 
zhuanlan.zhihu.com
501
vuejs-templates/webpack-simple: A simple Webpack + vue-loader setup for quick prototyping.
0 annotations
 
github.com
513
  • index.html # index.html template
  • App.vue # main app component
  • main.js # app entry file
Project Structure · GitBook
3 annotations
 
vuejs-templates.github.io
523
  • 颜色或纹理填充对象
  • 颜色取样
PixelStyle功能介绍-Mac苹果图片编辑软件
2 annotations
 
www.effectmatrix.com
513
  • ETL（提取转移负载）
  • 群集的更多信息
  • 36个ClickHouse节点
  • 每秒可以提供大约150个查询
  • 每秒服务约40次查询
  • 8192索引粒度
  • ReplicatedAggregatingMergeTree
  • SummingMergeTree
  • 聚合架构设计
  • 聚合架构设计
  • 该引擎指向非聚合请求表，并包含每个细分的精确聚合数据
  • 大量不同类型的指标
  • 存储超过100列
  • 原始（非聚合）请求日志存储在ClickHouse中1个月
  • 聚合应该适用于过去365天的任何时间范围
  • 许多不同的聚合
  • 引擎取代Kafka Go的消费者，因为它足够稳定，可以直接从Kafka摄取到ClickHouse
  • 水平可扩展性和容错性
  • 每秒6M请求的水平
  • 每秒1M的请求处理
  • ClickHouse构建新的和改进的管道的基础
  • 大规模数据基础架构
使用ClickHouse对每秒6百万次请求进行HTTP分析 - 发声吧
22 annotations
 
fashengba.com
515
  • 核心竞争力
  • 优势、劣势、机会和威胁
2017.2.16 《如何高效读懂一本书》 - 简书
2 annotations
 
www.jianshu.com
529
菜鸟教程在线编辑器
0 annotations
 
www.runoob.com
507
  • 大规模状态管理来
  • 客户端路由
  • 组件系统
  • 声明式渲染（视图模板引擎）
  • “渐进式”，其实就是Vue的使用方式
  • 逻辑控制负责联系视图板块和数据板块
  • 视图板块负责更改样式
  • 数据板块只负责数据部分
  • 负责的逻辑和巨大的开发量,是原生js无法完成
  • 渐进式框架
  • 建用户界面的渐进式框架
深入理解vue - 如是说 - 博客园
11 annotations
 
www.cnblogs.com
542
  • 还可以绑定到 DOM 结构
  • 不仅可以把数据绑定到 DOM 文本或 attribute
  • v-bind attribute 被称为指令
  • Vue.component
介绍 — Vue.js
4 annotations
 
cn.vuejs.org
504
  • 生命周期钩子的 this 上下文指向调用它的 Vue 实例
  • 只有当实例被创建时就已经存在于 data 中的 property 才是响应式的
  • 实例挂载到 DOM 并在数据变化时更新 DOM 等
  • 编译模板
  • 数据监听
  • Vue 实例在被创建时都要经过一系列的初始化过程
  • Vue 实例还暴露了一些有用的实例 property 与方法。它们都有前缀 $，以便与用户定义的 property 区分开来
  • data 对象中的所有的 property 加入到 Vue 的响应式系统中
  • Vue 组件都是 Vue 实例
Vue 实例 — Vue.js
9 annotations
 
cn.vuejs.org
575
  • Vue 提供了一个官方的 CLI，为单页面应用 (SPA) 快速搭建繁杂的脚手架
  • npm install vue
  • Vue 构建大型应用时推荐使用 NPM 安装[1]。NPM 能很好地和诸如 webpack 或 Browserify 模块打包器配合使用
  • NPM
  • CDN
  • 在 NPM 包的 dist/ 目录你将会找到很多不同的
  • CLI 工具假定用户对 Node.js 和相关构建工具有一定程度的了解
安装 — Vue.js
7 annotations
 
cn.vuejs.org
503
  • webpack可以看做是模块打包机，分析项目结构，找到JavaScript模块和一些浏览器不能直接运行的语言，类似Scss,JSX,Typescript等，并将其转换和打包为合适的格式供浏览器使用
webpack npm start和npm run server - 简书
1 annotations
 
www.jianshu.com
449
  • require 此文件之后，这个函数在其他代码中就可以使用了
  • exports对象的一个属性
  • 创建模块的入口文件
  • package.json
11 - 如何创建 Node.js 模块 | npm 中文文档
4 annotations
 
www.npmjs.cn
405