controlling the variance of request response times and not just worrying about maximizing queries per second
http://highscalability.com/blog/2010/11/4/facebook-at-13-million-queries-per-second-recommends-minimiz.html
Facebook gave a MySQL Tech Talk where they talked about many things MySQL, but one of the more subtle and interesting points was their focus on controlling the variance of request response times and not just worrying about maximizing queries per second.
But first the scalability porn. Facebook's OLTP performance numbers were as usual, quite dramatic:
- Query response times: 4ms reads, 5ms writes.
- Rows read per second: 450M peak
- Network bytes per second: 38GB peak
- Queries per second: 13M peak
- Rows changed per second: 3.5M peak
- InnoDB disk ops per second: 5.2M peak
Some thoughts on creating quality, not quantity:
- They don't care about average response times, instead, they want to minimize variance. Every click must be responded to quickly. The quality of service for each request matters.
- It's OK if a query is slow as long as it is always slow.
- They don't try to get the highest queries per second out of each machine. What is important is that the edge cases are not the bad.
- They figure out why the response time for the worst query is bad and then fix it.
- The performance community is often focussed on getting the highest queries per second. It's about making sure they have the best mix of IOPs available, cache size, and space.
To minimize variance they must be able notice, diagnose, and then fix problems:
- They measure how things work in operation. They can monitor at subsecond levels so they catch problems.
- Servers have miniature fractures in their performance which they call "stalls." They've built tools to find these.
- Dogpile collection. Every second it notices if something is wrong and ships it out for analysis.
- Poor man's profiler. Attach GDB to servers to know what's going on, they can see when stalls happen.
- Problems are usually counter-intuitive this can never happen type problems.
- Extending a table locks the entire instance.
- Flushing dirty pages was actually blocking.
- How statistics get sampled in InnoDB.
- Problems happen on medium-loaded systems too. Their systems aren't that loaded to ensure quality of service, yet the problems still happen.
- Analyze and understand every layer of the software stack to see how it performs at scale.
- Monitoring system monitors different aspects of performance so they can notice a change in performance, drill down to the host, then drill down to the query that might be causing the problem, then kill the query, and then trace it back to the source file where it occurred.
- They have a small team, so they make very specific changes to Linux and MySQL to support their use cases. Longer term changes are made by others.
Please watch the MySQL Tech Talk for more color and details.
controlling the variance of request response times and not just worrying about maximizing queries per second的更多相关文章
- request response
request 和 response 这两个对象是出现在service方法中.service方法是用来接收请求处理请求,完成响应的. 接受请求指的就是request对象 完成响应指的就 ...
- request \response 总结
request&response request 1.获得信息的方法 1> 获得请求首行信息的方法 *getMethod *getContextP ...
- track message forwards, avoiding request loops, and identifying the protocol capabilities of all senders along the request/response chain
https://www.w3.org/Protocols/rfc2616/rfc2616-sec9.html The TRACE method is used to invoke a remote, ...
- Ng Http Request/response格式转换
angular作为Single Page Application推荐的交互方式当然是基于json的ajax调用.但今天要说的是当你不幸工作在一个遗留或者不可控制的服务上,而这服务是基于非json提交方 ...
- 过滤器中的chain.doFilter(request,response)
Servlet中的过滤器Filter是实现了javax.servlet.Filter接口的服务器端程序,主要的用途是过滤字符编码.做一些业务逻辑判断等.其工作原理是,只要你在web.xml文件配置好要 ...
- 对chain.doFilter(request,response)的理解
他的作用是将请求转发给过滤器链上下一个对象.这里的“下”指的是哪里 ? 指的是下一个filter,如果没有filter那就是你请求的资源. 一般filter都是一个链,web.xml 里面配置了几个就 ...
- chain.doFilter(request,response)含义
过滤器的生命周期一般都要经过下面三个阶段: 初始化 当容器第一次加载该过滤器时,init() 方法将被调用.该类在这个方法中包含了一个指向 Filter Config 对象的引用.我们的过滤器实际上并 ...
- ZeroMQ之Request/Response (Java)
自己最开始是在cloud foundry中接触过消息服务器(nats),或者说是消息中间件,也算是初步知道了一个消息服务器对于分布式的网络系统的重要性,后来自己也曾想过在一些项目中使用它,尤其是在一些 ...
- 【转】对Django框架架构和Request/Response处理流程的分析
本文转载于疯狂的蚂蚁. 一. 处理过程的核心概念 如下图所示django的总览图,整体上把握以下django的组成: 核心在于中间件middleware,django所有的请求.返回都由中间件来完成. ...
随机推荐
- JavaScript 笔记 ( Prototype )
这阵子实在好忙 ( 这样说好像也不是一两个月了... ),然后因为工作伙伴都是 JavaScript 神之等级的工程师,从中也学到不少知识,毕竟就是要和强者工作才会成长呀!为了想好好瞭解他们写的程式码 ...
- BestCoder#16 A-Revenge of Segment Tree
Revenge of Segment Tree Problem Description In computer science, a segment tree is a tree data struc ...
- 不刷新改变URL: pushState + Ajax
如果你玩过Google+,看到过YouTube的新界面,便会体验到这个HTML5的新功能.使用pushState + Ajax(pjax),可以实现网页的ajax加载,同时又能完成URL的改变而没有网 ...
- BZOJ 2049 & LCT又一模板
题意: 自己看...加边删边问联通... SOL: 就加了一个findroot而已... 然而时间还是惨不忍睹...优化全开也才1700ms...膜seter... Code: /*========= ...
- UVa 11324 & 强联通分量+DP
题意: 一张无向图,求点集使其中任意两点可到达. SOL: 强联通分量中的点要么不选要么全都选,然后缩点DAG+DP 记录一下思路,不想写了...代码满天飞.
- UVa 10806 & 费用流+意识流...
题意: 一张无向图,求两条没有重复的从S到T的路径. SOL: 网络流为什么屌呢..因为网络流的容量,流量,费用能对许许多多的问题进行相应的转化,然后它就非常的屌. 对于这道题呢,不是要没有重复吗?不 ...
- HDU 2851 (最短路)
题目链接:http://acm.hdu.edu.cn/showproblem.php?pid=2851 题目大意:给出N条路径,M个终点(是路径的编号) .重合的路径才算连通的,且路径是单向的.每条路 ...
- 【BZOJ】2697: 特技飞行
题意 \(k(1 \le k \le 300)\)种物品,价值分别为\(c_i(0 \le c_i \le 1000)\).有\(n(1 \le n \le 1000)\)分钟,每分钟可以选择一个物品 ...
- Android -- ids.xml文件的使用
1.当我们在使用控件的时候,有的时候不可避免的要为控件声明id ,而分散在各个xml文件中,有时候查找起来又不是特别的方便 ,因此,安卓为我们提供了ids.xml 文件,保存在res -->va ...
- Android -- 闹钟服务的使用(单次闹钟)
1. 效果图