search
HomeDatabaseMysql TutorialQihoo360 Atlas MySQL Proxy测试小结

Qihoo360将他们改造后的MySQL Proxy项目开源了,至于为什么起名Atlas就不清楚了,项目地址:https://github.com/Qihoo360/Atlas。我2008年曾测试过官方版本的MySQL Proxy,主要是看中其连接池以及读写分离功能,不过当时的版本效率实在太差,后面就没再关注

Qihoo360将他们改造后的MySQL Proxy项目开源了,至于为什么起名Atlas就不清楚了,项目地址:https://github.com/Qihoo360/Atlas。我2008年曾测试过官方版本的MySQL Proxy,主要是看中其连接池以及读写分离功能,不过当时的版本效率实在太差,后面就没再关注了。这几天对Qihoo360 Atlas做了下测试,下面是测试结果。

  • 环境准备

服务器端:

测试机 DELL PE R710
CPU E5620? @ 2.40GHz(4 core, 8 threads) * 2
内存 24G
RAID卡 PERC H700 Integrated, 512MB, BBU, 12.10.1-0001
系统 Red Hat Enterprise Linux Server release 6.4 (Santiago)
内核 2.6.32-358.el6.x86_64 #1 SMP
raid级别 raid 5(10K RPM SAS 300G * 6)
文件系统 xfs
硬盘 10K RPM SAS 300G * 6

Proxy代理层端:

测试机 HP DL360 G5
CPU E5405?@ 2.00GHz(4 core, 4 threads) * 2
内存 16G
RAID卡 P400i 256MB, BBU,?4.12
系统 Red Hat Enterprise Linux Server release 6.1 (Santiago)
内核 2.6.32-131.0.15.el6.x86_64?#1 SMP
raid级别 raid 0(10K RPM SAS 146G * 2)
文件系统 ext4
硬盘 10K RPM SAS 146G* 2
  • MySQL及Atlas关键配置
[yejr@imysql.com]# cat /etc/my.cnf
max_connections = 2048
max_connect_errors = 100000
[yejr@imysql.com]# cat /usr/local/mysql-proxy/conf/13306.cnf
event-threads = 16
min-idle-connections = 768
  • 测试方案

1. 采用tpcc-mysql进行压测,但由于Atlas不支持PREPARE,未遂;

2. 采用sysbench进行压测,但由于Atlas当前版本对长连接支持不佳,出现大量的connection lost告警,测试结果几乎无效,忽略;

3. 采用开发者提供的一个简易C程序,进行并发多线程短连接测试,虽然有部分connection lost告警,但比用sysbench时少多了,结果有效;

开发者提供的C程序,若有需要可向开发者提出,由于未得到授权,我不方便将其放上来供下载。

测试模式:

测试程序并发数:32、64、128、192、256、320、384、448、512、576、640,每线程完成10000次请求,每次请求都是随机的SELECT、INSERT、UPDATE,每次UPDATE都是基于主键条件,INSERT就不用说了,SELECT有4种随机模式:

a)?/*master*/SELECT * FROM mysqlslap.t1 WHERE /*waht*/id in (N1, N2, N3)

b)?SELECT * FROM mysqlslap.t1 WHERE id = N

c)?SELECT * FROM mysqlslap.t1 WHERE id in (N1, N2) limit 1

d)?/*master*/SELECT * FROM /*test*/ mysqlslap.t1 WHERE id = %d

也就是SELECT请求会模拟强制读MASTER,或者让Atlas自动分配。

  • 测试脚本

测试脚本很简单:

[yejr@imysql.com]# cat Atlas_benchmark.sh
#!/bin/bash
MYSERVER=10.0.0.1
PORT=13306
export LD_LIBRARY_PATH=/usr/local/mysql/lib/mysql
for THREAD in 32 64 128 192 256 320 384 448 512 576 640
do
#记录日志
exec 3>&1 4>&2 1>> check_1m3s_1server_1proxy_${THREAD}_${RANDOM}.log 2>&1
count=1
max=5
#每种并发模式都进行5轮测试,最后取结果平均值
while [ $count -le ${max} ]
do
echo "./check -t short -c ${THREAD} -h ${MYSERVER} -P ${PORT} -u proxy -p proxy -n 10000"
./check -t short -c ${THREAD} -h ${MYSERVER} -P ${PORT}-u proxy -p proxy -n 10000
count=`expr ${count} + 1`
#每次测试完,都会暂停60秒,让数据库歇歇 :)
if [ ${count} -lt ${max} ] ; then
 sleep 60
fi
done
done

测试完后对结果进行整理汇总即可。

  • 测试结果

测试成功次数

Atlas Proxy并发测试成功次数 - 20130911

Atlas Proxy并发测试成功次数 – 20130911

测试成功率

Atlas Proxy并发测试成功率 - 20130911

Atlas Proxy并发测试成功率 – 20130911

  • 结论及建议

Qihoo360开源的精神值得学习,不过目前来看Atlas还有一定提升空间,不妨等它更加稳定并且能支持更多特性后再上线使用不迟。

但是对那些苦于无法将slave从库资源利用起来的同学们来说,在前端加一层Atlas倒是很不错的选择,毕竟其他可选的产品太少了,只要程序中不用到特殊的SQL一般也不会有大问题,注意提高程序对数据的容错性即可。

Statement
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
Explain the role of InnoDB redo logs and undo logs.Explain the role of InnoDB redo logs and undo logs.Apr 15, 2025 am 12:16 AM

InnoDB uses redologs and undologs to ensure data consistency and reliability. 1.redologs record data page modification to ensure crash recovery and transaction persistence. 2.undologs records the original data value and supports transaction rollback and MVCC.

What are the key metrics to look for in an EXPLAIN output (type, key, rows, Extra)?What are the key metrics to look for in an EXPLAIN output (type, key, rows, Extra)?Apr 15, 2025 am 12:15 AM

Key metrics for EXPLAIN commands include type, key, rows, and Extra. 1) The type reflects the access type of the query. The higher the value, the higher the efficiency, such as const is better than ALL. 2) The key displays the index used, and NULL indicates no index. 3) rows estimates the number of scanned rows, affecting query performance. 4) Extra provides additional information, such as Usingfilesort prompts that it needs to be optimized.

What is the Using temporary status in EXPLAIN and how to avoid it?What is the Using temporary status in EXPLAIN and how to avoid it?Apr 15, 2025 am 12:14 AM

Usingtemporary indicates that the need to create temporary tables in MySQL queries, which are commonly found in ORDERBY using DISTINCT, GROUPBY, or non-indexed columns. You can avoid the occurrence of indexes and rewrite queries and improve query performance. Specifically, when Usingtemporary appears in EXPLAIN output, it means that MySQL needs to create temporary tables to handle queries. This usually occurs when: 1) deduplication or grouping when using DISTINCT or GROUPBY; 2) sort when ORDERBY contains non-index columns; 3) use complex subquery or join operations. Optimization methods include: 1) ORDERBY and GROUPB

Describe the different SQL transaction isolation levels (Read Uncommitted, Read Committed, Repeatable Read, Serializable) and their implications in MySQL/InnoDB.Describe the different SQL transaction isolation levels (Read Uncommitted, Read Committed, Repeatable Read, Serializable) and their implications in MySQL/InnoDB.Apr 15, 2025 am 12:11 AM

MySQL/InnoDB supports four transaction isolation levels: ReadUncommitted, ReadCommitted, RepeatableRead and Serializable. 1.ReadUncommitted allows reading of uncommitted data, which may cause dirty reading. 2. ReadCommitted avoids dirty reading, but non-repeatable reading may occur. 3.RepeatableRead is the default level, avoiding dirty reading and non-repeatable reading, but phantom reading may occur. 4. Serializable avoids all concurrency problems but reduces concurrency. Choosing the appropriate isolation level requires balancing data consistency and performance requirements.

MySQL vs. Other Databases: Comparing the OptionsMySQL vs. Other Databases: Comparing the OptionsApr 15, 2025 am 12:08 AM

MySQL is suitable for web applications and content management systems and is popular for its open source, high performance and ease of use. 1) Compared with PostgreSQL, MySQL performs better in simple queries and high concurrent read operations. 2) Compared with Oracle, MySQL is more popular among small and medium-sized enterprises because of its open source and low cost. 3) Compared with Microsoft SQL Server, MySQL is more suitable for cross-platform applications. 4) Unlike MongoDB, MySQL is more suitable for structured data and transaction processing.

How does MySQL index cardinality affect query performance?How does MySQL index cardinality affect query performance?Apr 14, 2025 am 12:18 AM

MySQL index cardinality has a significant impact on query performance: 1. High cardinality index can more effectively narrow the data range and improve query efficiency; 2. Low cardinality index may lead to full table scanning and reduce query performance; 3. In joint index, high cardinality sequences should be placed in front to optimize query.

MySQL: Resources and Tutorials for New UsersMySQL: Resources and Tutorials for New UsersApr 14, 2025 am 12:16 AM

The MySQL learning path includes basic knowledge, core concepts, usage examples, and optimization techniques. 1) Understand basic concepts such as tables, rows, columns, and SQL queries. 2) Learn the definition, working principles and advantages of MySQL. 3) Master basic CRUD operations and advanced usage, such as indexes and stored procedures. 4) Familiar with common error debugging and performance optimization suggestions, such as rational use of indexes and optimization queries. Through these steps, you will have a full grasp of the use and optimization of MySQL.

Real-World MySQL: Examples and Use CasesReal-World MySQL: Examples and Use CasesApr 14, 2025 am 12:15 AM

MySQL's real-world applications include basic database design and complex query optimization. 1) Basic usage: used to store and manage user data, such as inserting, querying, updating and deleting user information. 2) Advanced usage: Handle complex business logic, such as order and inventory management of e-commerce platforms. 3) Performance optimization: Improve performance by rationally using indexes, partition tables and query caches.

See all articles

Hot AI Tools

Undresser.AI Undress

Undresser.AI Undress

AI-powered app for creating realistic nude photos

AI Clothes Remover

AI Clothes Remover

Online AI tool for removing clothes from photos.

Undress AI Tool

Undress AI Tool

Undress images for free

Clothoff.io

Clothoff.io

AI clothes remover

AI Hentai Generator

AI Hentai Generator

Generate AI Hentai for free.

Hot Article

R.E.P.O. Energy Crystals Explained and What They Do (Yellow Crystal)
4 weeks agoBy尊渡假赌尊渡假赌尊渡假赌
R.E.P.O. Best Graphic Settings
4 weeks agoBy尊渡假赌尊渡假赌尊渡假赌
R.E.P.O. How to Fix Audio if You Can't Hear Anyone
4 weeks agoBy尊渡假赌尊渡假赌尊渡假赌
WWE 2K25: How To Unlock Everything In MyRise
1 months agoBy尊渡假赌尊渡假赌尊渡假赌

Hot Tools

Zend Studio 13.0.1

Zend Studio 13.0.1

Powerful PHP integrated development environment

DVWA

DVWA

Damn Vulnerable Web App (DVWA) is a PHP/MySQL web application that is very vulnerable. Its main goals are to be an aid for security professionals to test their skills and tools in a legal environment, to help web developers better understand the process of securing web applications, and to help teachers/students teach/learn in a classroom environment Web application security. The goal of DVWA is to practice some of the most common web vulnerabilities through a simple and straightforward interface, with varying degrees of difficulty. Please note that this software

EditPlus Chinese cracked version

EditPlus Chinese cracked version

Small size, syntax highlighting, does not support code prompt function

SublimeText3 Mac version

SublimeText3 Mac version

God-level code editing software (SublimeText3)

Safe Exam Browser

Safe Exam Browser

Safe Exam Browser is a secure browser environment for taking online exams securely. This software turns any computer into a secure workstation. It controls access to any utility and prevents students from using unauthorized resources.