https://logdevice.io/blog/2018/09/12/open-sourcing-announcement.html

September 12, 2018

 

We are excited to announce that today we are open-sourcing LogDevice! The code is now available on GitHub under the permissive 3-clause BSD license.

LogDevice is a distributed data store for sequential data originally developed within Facebook. This 2017 blog post in the Facebook Engineering blog explains the basic components of LogDevice. You can also refer to the recording of a presentation from the @Scale conference for more detail.

LogDevice is widely deployed for use at Facebook. Existing use cases include:

  • Stream processing pipelines
  • Distribution of index updates in large distributed databases
  • Machine learning pipelines
  • Replication pipelines
  • Durable reliable task queues

Over the past few years we have built a large number of internal tools to manage LogDevice clusters. We are working on slowly shifting some of that automation to be shipped in the open-source version as well. For this initial release, we are including a bare-bones toolset that forms the core of LogDevice operational infrastructure. In a large-scale production environment it will need to be supplemented by other tools for tasks like monitoring and cluster topology management.

LogDevice comes with a command-line administration tool called LDShell (ldshell). LDShell is the primary management tool that we use here at Facebook as well. Stay tuned as we add more commands and automation to it. This is one of the areas where we expect support from the open-source community.

We are excited to see your contributions to LogDevice, especially those related to integration with cloud computing environments, monitoring systems, and configuration management. Getting LogDevice to run on Kubernetes is a good starting point if you are interested in helping.

To get started with LogDevice, dive into the overview of the system to learn how it works, or check out the code on GitHub

By Ahmed Soliman and Andrejs Krasilnikovs, LogDevice team.

 

Open-sourcing LogDevice, a distributed data store for sequential data的更多相关文章

  1. Bigtable:A Distributed Storage System for Strctured Data

    2006 年10 月Google 发布三架马车之一的<Bigtable:A Distributed Storage System for Strctured Data>论文之后,Power ...

  2. 分布式存储——Build up a High Availability Distributed Key-Value Store

    原文链接 Preface There are many awesome and powerful distributed NoSQL in the world, like Couchbase, Mon ...

  3. 设置 Ext.data.Store 传参的请求方式

    设置 Ext.data.Store 传参的请求方式 1.extjs 给怎么给panel设背景色 设置bodyStyle:'background:#ffc;padding:10px;', var res ...

  4. Ext.data.Store添加动态参数

    多条件查询页面的参数都是动态的,并且我们通常还会有默认加载页面.此时,动态添加参数非常重要,其中baseparam是解决问题的关键. @ 将查询条件定义为一个全局变量 var param_01 = & ...

  5. 转: Ext.data.Store 修改Post请求

    Extjs 4.0版本 var Store = Ext.create('Ext.data.Store', { pageSize: pageSize, model: 'Ext.data.Model名称' ...

  6. 对于Ext.data.Store 介紹 与总结,以及对以前代码的重构与优化

    对于Ext.data.Store 一直不是很了解,不知道他到底是干嘛的有哪些用处,在实际开发中也由于不了解也走了不少弯路, store是一个为Ext器件提供record对象的存储容器,行为和属性都很象 ...

  7. Extjs 项目中常用的小技巧,也许你用得着(5)--设置 Ext.data.Store 传参的请求方式

    1.extjs 给怎么给panel设背景色 设置bodyStyle:'background:#ffc;padding:10px;', var resultsPanel = Ext.create('Ex ...

  8. 【ArcGIS】ArcGIS Data Store配置

    一.错误提示 Unable to configure the ArcGIS Data Store with the GIS Server. Please make sure that the GIS ...

  9. sencha touch carousel 扩展 CardList 可绑定data/store

    扩展代码: /* *扩展carousel *通过data,tpl,store配置数据 */ Ext.define('ux.CardList', { extend: 'Ext.carousel.Caro ...

随机推荐

  1. 源码分析——迁移学习Inception V3网络重训练实现图片分类

    1. 前言 近些年来,随着以卷积神经网络(CNN)为代表的深度学习在图像识别领域的突破,越来越多的图像识别算法不断涌现.在去年,我们初步成功尝试了图像识别在测试领域的应用:将网站样式错乱问题.无线领域 ...

  2. UVA 11389 The Bus Driver Problem 贪心水题

    题目链接:UVA - 11389 题意描述:有n个司机,n个早班路线和n个晚班路线,给每个司机安排一个早班路线和一个晚班路线,使得每个早班路线和晚班路线只属于一个司机.如果一个司机早班和晚班总的驾驶时 ...

  3. 【spring cloud】【spring boot】项目启动报错:Cannot determine embedded database driver class for database type NONE

    解决参考文章:https://blog.csdn.net/hengyunabc/article/details/78762097 spring boot启动报错如下: Error starting A ...

  4. 【IntelliJ Idea】idea下hibernate反向生成工具,根据数据表生成实体

    idea插件很齐全,不像ecplise一样.所以直接来步骤吧: 1.选择项目,右键-->Add Frameworks Support-->勾选Hibernate-->勾选Import ...

  5. Backup and Recovery Basics2

    1.6.Automatic Disk-Based Backup and Recovery: The Flash Recovery Area 创建不同备份和恢复文件的组件对每一个文件系统的大小没有不论什 ...

  6. UNP学习笔记(第十八章 路由套接字)

    路由套接字上支持3种类型的操作 1). 进程能通过写路由套接字向内核发消息. 2). 进程能通过路由套接字从内核读消息. 3). 进程可以用sysctl函数得到路由表或列出所有已配置的接口. 数据链路 ...

  7. 【Python】python3中urllib爬虫开发

    以下是三种方法 ①First Method 最简单的方法 ②添加data,http header 使用Request对象 ③CookieJar import urllib.request from h ...

  8. 世纪怎么换算成具体的年份?eg:19世纪60年代=19??年?

    http://zhidao.baidu.com/question/339742625.html&__bd_tkn__=6ab5183c226b84031b08b849ecac35b396039 ...

  9. 用命令行执行ROBOT FRAMEWORK

    除了在ride中执行用例,我们也可以通过命令行的形式执行用例. 1.执行一整个项目 pybot+项目路径 2.执行某个测试集 pybot+测试集的路径 3.执行某个测试集里面的某个用例 pybot - ...

  10. Realm多线程中的那些坑...

    个人在开发中遇到的一些小坑... 可能会持续更新... 1.RealmObject自带线程保护功能.仅仅能在创建它的线程中訪问.在子线程中不能訪问. 也就是说.假设你在主线程中new了一个RealmO ...