本帖最后由 ylky_2000 于 2015-5-20 09:34 编辑
2015-05-20 9:30
1)首先请做一道选择题:A坚守X86,B坚守POWER,C从X86迁移到POWER,D从POWER迁移到X86,E同时使用X86和POWER。请选择并具体说明你的理由。 初期我会选择E,毕竟系统稳定运行是头等大事,在没生产环境中应用经验和排错经验的时候可以慢慢来。同时可以比较x86与power的不同点,相信ibm放弃x86,主推power,肯定有原因,我们只是需要在生成环境中验证并享受这个优势即可。后期power确实强劲,稳定,会选C。
2)如果你曾经在X86和非X86平台之间迁移过应用,请具体分享一下当时的经历和心得?比如,当时为什么要迁移?在迁移的过程中碰到什么问题?结合下面这些因素,想想当时是怎么考虑的:业务需求、成本、稳定安全性、扩展性、系统和应用兼容性、是否被厂商绑定、服务等。 实际的生成环境的经验没有,不过测试过,下面有一个我当时的测试报告,供参考,迁移测试过程顺利。讨论。测试的应用软件没有问题,大家放心使用,也欢迎提供更多的测试案例,补充丰富。
3)结合IBM最近在推的Linux On POWER,和英特尔刚刚发布的超级处理器至强E7 V3,两大厂商攻守之间,你更看好谁?是英特尔至强E7继续攻城拔寨,吸引诸如银行、电信等客户进一步将核心业务转向X86,还是IBM凭借Linux On POWER在守住阵脚之后绝地反击,比如吸引互联网厂商转投POWER怀抱?元芳,你怎么看? 建议搞一些事件营销。比如12306购票网站,提供给他们解决他们的实际问题,x86抗不住,power抗住了。就是很好的证明。 阿里云不是跟12306合作了嘛,把票查询都放在阿里云上了,再看看阿里云从12306合作后的市场份额吧。绝对可行。就当是打广告了吧。
4)今天市场上各种开放平台不断涌现,开源技术大行其道,封闭的IT架构似乎已经走入死胡同。那么,请谈谈你期待这些开放平台能带来什么具体的价值?比如灵活性、低成本、可选择性、创新与合作机会等。 开放意味着,灵活、可选、通过开放的接口集成和个性定制化开发。
2015-05-20早上8点多
发一份 最近测试过的power架构的云主机报告
一、参数收集
1、网络部分
1.1ip地址
182.18.34.23 只有一个网卡。 Mac地址FA:33:E3:87:F0:20
1.2网卡型号
[root@t3 ~]# lspci -vv 没有显示。
1.3网卡速率
[root@t3 ~]# mii-tool -v eth0
SIOCGMIIPHY on 'eth0' failed: Operation not supported
1.4路由表
[root@t3 ~]# route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
182.18.34.0 0.0.0.0 255.255.255.192 U 0 0 0 eth0
169.254.0.0 0.0.0.0 255.255.0.0 U 1002 0 0 eth0
0.0.0.0 182.18.34.1 0.0.0.0 UG 0 0 0 eth0
1.5iptables表
[root@t3 ~]# iptables -L -n
Chain INPUT (policy ACCEPT)
target prot opt source destination
SST-INPUT all -- 0.0.0.0/0 0.0.0.0/0
Chain FORWARD (policy ACCEPT)
target prot opt source destination
Chain OUTPUT (policy ACCEPT)
target prot opt source destination
SST-OUTPUT all -- 0.0.0.0/0 0.0.0.0/0
Chain SST-INPUT (1 references)
target prot opt source destination
ACCEPT tcp -- 0.0.0.0/0 0.0.0.0/0 tcp dpt:6060
Chain SST-OUTPUT (1 references)
target prot opt source destination
[root@t3 ~]# cat /etc/sysconfig/iptables
# Generated by iptables-save v1.4.7 on Mon Mar 16 09:23:43 2015
*filter
:INPUT ACCEPT [0:0]
:FORWARD ACCEPT [0:0]
:OUTPUT ACCEPT [0:0]
:SST-INPUT - [0:0]
:SST-OUTPUT - [0:0]
-A INPUT -j SST-INPUT
-A OUTPUT -j SST-OUTPUT
-A SST-INPUT -p tcp -m tcp --dport 6060 -j ACCEPT
COMMIT
# Completed on Mon Mar 16 09:23:43 2015
# Generated by iptables-save v1.4.7 on Mon Mar 16 09:23:43 2015
*nat
REROUTING ACCEPT [0:0]
OSTROUTING ACCEPT [0:0]
:OUTPUT ACCEPT [0:0]
:SST-POSTROUTING - [0:0]
:SST-PREROUTING - [0:0]
-A PREROUTING -j SST-PREROUTING
-A POSTROUTING -j SST-POSTROUTING
COMMIT
# Completed on Mon Mar 16 09:23:43 2015
# Generated by iptables-save v1.4.7 on Mon Mar 16 09:23:43 2015
*mangle
REROUTING ACCEPT [0:0]
:INPUT ACCEPT [0:0]
:FORWARD ACCEPT [0:0]
:OUTPUT ACCEPT [0:0]
OSTROUTING ACCEPT [0:0]
COMMIT
# Completed on Mon Mar 16 09:23:43 2015
这个6060端口不知道是什么东西。
2、系统硬件方面
2.1cpu
[root@t3 ~]# cat /proc/cpuinfo
processor : 0
cpu : POWER7 ibm最新的是power8结构,测试机是power7架构也很强大。
machine : CHRP IBM,8246-L2T。
CPU maps initialized for 4 threads per core
PERCPU: Embedded 2 pages/cpu @c000000001f00000 s92840 r0 d38232 u131072
Brought up 8 CPUs
Node 0 CPUs: 0-7
[root@t3 ~]# cat /proc/cpuinfo |grep "physical id"|sort|uniq |wc -l
0
只有一个cpu、逻辑个数(process)为8个。
这里跟提供的有出入,不知道是我测试的问题还是?
2.2硬盘
分区情况
[root@t3 ~]# df -ah
Filesystem Size Used Avail Use% Mounted on
/dev/sda4 39G 9.2G 28G 26% /
proc 0 0 0 - /proc
sysfs 0 0 0 - /sys
devpts 0 0 0 - /dev/pts
tmpfs 2.0G 0 2.0G 0% /dev/shm
/dev/sda2 97M 42M 51M 46% /boot
none 0 0 0 - /proc/sys/fs/binfmt_misc
/root/ilanni/CentOS-6.6-i386-minimal.iso 339M 339M 0 100% /iso
/dev/sr0 3.2G 3.2G 0 100% /mnt
硬盘空间
[root@t3 ~]# fdisk -l
跟提供的环境一致
Disk /dev/sda: 42.9 GB, 42949672960 bytes
64 heads, 32 sectors/track, 40960 cylinders
Units = cylinders of 2048 * 512 = 1048576 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x6f220300
Device Boot Start End Blocks Id System
/dev/sda1 * 2 9 8192 41 PPC PReP Boot
Partition 1 does not end on cylinder boundary.
/dev/sda2 10 109 102400 83 Linux
Partition 2 does not end on cylinder boundary.
/dev/sda3 110 1133 1048576 82 Linux swap / Solaris
Partition 3 does not end on cylinder boundary.
/dev/sda4 1134 40960 40782848 83 Linux
Partition 4 does not end on cylinder boundary.
2.3内存
4g与提供的一致
[root@t3 ~]# free -m
total used free shared buffers cached
Mem: 4034 2686 1348 0 226 2007
2.4操作系统
[root@t3 ~]# lsb_release -a
LSB Version: :base-4.0-noarch:base-4.0-ppc64:core-4.0-noarch:core-4.0-ppc64:graphics-4.0-noarch:graphics-4.0-ppc64 rinting-4.0-noarch rinting-4.0-ppc64
Distributor ID: RedHatEnterpriseServer
Description: Red Hat Enterprise Linux Server release 6.5 (Santiago)
Release: 6.5
Codename: Santiago
二、性能测试
2.1 硬盘性能测试
/dev/sda:
Timing cached reads: 37182 MB in 1.99 seconds = 18720.14 MB/sec
Timing buffered disk reads: 542 MB in 3.01 seconds = 180.18 MB/sec
写磁盘速度测试
10.28s左右生成了一个2.5GB的文件,速度239MB/s
[root@t3 ~]# time dd if=/dev/zero of=/test.dbf bs=8k count=300000
300000+0 records in
300000+0 records out
2457600000 bytes (2.5 GB) copied, 10.2835 s, 239 MB/s
real 0m10.292s
user 0m0.092s
sys 0m1.850s
读磁盘速度测试
[root@t3 ~]# df -m
Filesystem 1M-blocks Used Available Use% Mounted on
/dev/sda4 39202 11724 25487 32% /
tmpfs 2018 0 2018 0% /dev/shm
/dev/sda2 97 42 51 46% /boot
/root/ilanni/CentOS-6.6-i386-minimal.iso 339 339 0 100% /iso
/dev/sr0 3193 3193 0 100% /mnt
[root@t3 ~]# time dd if=/dev/sda4 of=/dev/null bs=8k
5097856+0 records in
5097856+0 records out
41761636352 bytes (42 GB) copied, 223.744 s, 187 MB/s
real 3m43.745s
user 0m1.454s
sys 0m18.486s
2.2网络延时测试
查询的 IP:182.18.34.23 来自:北京市
GeoIP: Beijing, China
Beijing HuaxiaYakue Network Technology Co., Ltd.
2.3网络带宽测试
[root@t3 ~]# wget https://github.com/sivel/speedte ... er/speedtest_cli.py
[root@t3 ~]# ./speedtest-cli --help
Usage: speedtest-cli [options]
Command line interface for testing internet bandwidth using speedtest.net.
--------------------------------------------------------------------------
https://github.com/sivel/speedtest-cli
Options:
-h, --help show this help message and exit
--bytes Display values in bytes instead of bits. Does not affect
the image generated by --share
--share Generate and provide a URL to the speedtest.net share
results image
--simple Suppress verbose output, only show basic information
--list Display a list of speedtest.net servers sorted by
distance
--server=SERVER Specify a server ID to test against
--mini=MINI URL of the Speedtest Mini server
--source=SOURCE Source IP address to bind to
--timeout=TIMEOUT HTTP timeout in seconds. Default 10
--version Show the version number and exit
[root@t3 ~]# ./speedtest-cli --list |grep China
测试到北京电信带宽
[root@t3 ~]# ./speedtest-cli --server=4751 --share
Retrieving speedtest.net configuration...
Retrieving speedtest.net server list...
Testing from Beijing yiantianxia Network (182.18.34.23)...
Hosted by Beijing Telecom (Beijing) [8.08 km]: 2.759 ms
Testing download speed........................................
Download: 60.69 Mbit/s
Testing upload speed..................................................
Upload: 65.61 Mbit/s
Share results: https://www.speedtest.net/result/4310070563.png
测试到中国移动带宽
选了3个测试数据都为0,感觉跟移动之间互通有点问题。
测试到中国联通带宽
[root@t3 ~]# ./speedtest-cli --server=3891 --share
Retrieving speedtest.net configuration...
Retrieving speedtest.net server list...
Testing from Beijing yiantianxia Network (182.18.34.23)...
Hosted by Unicom-GZ (Guangzhou) [1893.48 km]: 52.615 ms
Testing download speed........................................
Download: 71.17 Mbit/s
Testing upload speed..................................................
Upload: 31.63 Mbit/s
Share results: https://www.speedtest.net/result/4310086747.png
三、常见应用测试
3.1web应用测试
3.1.1启动apache
已经安装好2.2.15版本的apache
[root@t3 ~]# rpm -qa|grep httpd
httpd-2.2.15-29.el6_4.ppc64
httpd-tools-2.2.15-29.el6_4.ppc64
启动apache 查看进程已经启动了
[root@t3 ~]# ps aux|grep httpd
root 1513 0.0 0.3 197120 13952 ? Sl Mar16 3:31 python /opt/ibm/sst/tools/httpserver/simplehttpd -d -p /var/run/uppd.pid -c /etc/upp/conf.d/
root 22382 0.0 0.7 115648 33024 ? Ss 01:57 0:00 /usr/sbin/httpd
apache 22386 0.0 0.4 115648 19648 ? S 01:57 0:00 /usr/sbin/httpd
apache 22387 0.0 0.4 115648 19648 ? S 01:57 0:00 /usr/sbin/httpd
apache 22388 0.0 0.4 115648 19648 ? S 01:57 0:00 /usr/sbin/httpd
apache 22389 0.0 0.4 115648 19648 ? S 01:57 0:00 /usr/sbin/httpd
apache 22390 0.0 0.4 115648 19648 ? S 01:57 0:00 /usr/sbin/httpd
apache 22391 0.0 0.4 115648 19648 ? S 01:57 0:00 /usr/sbin/httpd
apache 22392 0.0 0.4 115648 19648 ? S 01:57 0:00 /usr/sbin/httpd
apache 22393 0.0 0.4 115648 19648 ? S 01:57 0:00 /usr/sbin/httpd
root 22415 0.0 0.0 104832 2816 pts/3 S+ 01:58 0:00 grep httpd
默认80端口
http://182.18.34.23/
出来了。。温馨的apache页面
看看php信息
http://182.18.34.23/1.php 5.3.3版本
3.1.2压力测试
1、100人并发测试 通过
[root@t3 html]# ab -n 4000 -c 100 http://182.18.34.23/1.php
This is ApacheBench, Version 2.3 <$Revision: 655654 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/
Benchmarking 182.18.34.23 (be patient)
Completed 400 requests
Completed 800 requests
Completed 1200 requests
Completed 1600 requests
Completed 2000 requests
Completed 2400 requests
Completed 2800 requests
Completed 3200 requests
Completed 3600 requests
Completed 4000 requests
Finished 4000 requests
Server Software: Apache/2.2.15
Server Hostname: 182.18.34.23
Server Port: 80
Document Path: /1.php
Document Length: 47303 bytes
Concurrency Level: 100
Time taken for tests: 2.073 seconds
Complete requests: 4000
Failed requests: 0
Write errors: 0
Total transferred: 189904000 bytes
HTML transferred: 189212000 bytes
Requests per second: 1929.78 [#/sec] (mean)
Time per request: 51.819 [ms] (mean)
Time per request: 0.518 [ms] (mean, across all concurrent requests)
Transfer rate: 89470.81 [Kbytes/sec] received
Connection Times (ms)
min mean[+/-sd] median max
Connect: 3 24 11.0 24 45
Processing: 6 28 11.1 27 48
Waiting: 1 22 11.9 22 45
Total: 46 52 1.7 52 54
Percentage of the requests served within a certain time (ms)
50% 52
66% 53
75% 53
80% 53
90% 53
95% 54
98% 54
99% 54
100% 54 (longest request)
[root@t3 ~]# vmstat -n 1 cpu变化情况 内存基本没有影响
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu-----
r b swpd free buff cache si so bi bo in cs us sy id wa st
0 0 0 270528 1589888 1105472 0 0 2 2 0 0 0 0 100 0 0
0 0 0 273536 1589888 1105472 0 0 0 0 4 27 0 0 100 0 0
4 0 0 270208 1589888 1105472 0 0 0 0 624 1455 5 1 94 0 0
5 0 0 266432 1589888 1106048 0 0 0 0 5216 11809 36 8 56 0 0
0 0 0 271360 1589888 1107072 0 0 0 0 5180 11710 34 8 59 0 0
0 0 0 276736 1589888 1107200 0 0 0 0 4 31 0 0 100 0 0
0 0 0 284032 1589888 1107584 0 0 0 36 9 46 0 0 100 0 0
3.2数据库应用测试
3.2.1启动进入mysql数据库
[root@t3 html]# mysql -uroot -p
Enter password:
Welcome to the MySQL monitor. Commands end with ; or \g.
Your MySQL connection id is 2
Server version: 5.1.71 Source distribution
Copyright (c) 2000, 2013, Oracle and/or its affiliates. All rights reserved.
Oracle is a registered trademark of Oracle Corporation and/or its
affiliates. Other names may be trademarks of their respective
owners.
Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.
mysql>
mysql> show databases;
+--------------------+
| Database |
+--------------------+
| information_schema |
| mysql |
| test |
+--------------------+
3 rows in set (0.00 sec)
3.2.2性能测试
迭代测试
[root@t3 ~]# mysqlslap -aT -i 10 -c 200
Benchmark
Average number of seconds to run all queries: 0.239 seconds
Minimum number of seconds to run all queries: 0.224 seconds
Maximum number of seconds to run all queries: 0.247 seconds
Number of clients running queries: 200
Average number of queries per client: 0
User time 2.90, System time 1.12
Maximum resident set size 38400, Integral resident set size 0
Non-physical pagefaults 3322, Physical pagefaults 5, Swaps 0
Blocks in 2008 out 0, Messages in 0 out 0, Signals 0
Voluntary context switches 82916, Involuntary context switches 636
测试50,100个并发 1000次查询
[root@t3 ~]# mysqlslap -a --concurrency=50,100 --number-of-queries 1000 --debug-info -uroot -p
Enter password:
Benchmark
Average number of seconds to run all queries: 0.062 seconds
Minimum number of seconds to run all queries: 0.062 seconds
Maximum number of seconds to run all queries: 0.062 seconds
Number of clients running queries: 50
Average number of queries per client: 20
Benchmark
Average number of seconds to run all queries: 0.067 seconds
Minimum number of seconds to run all queries: 0.067 seconds
Maximum number of seconds to run all queries: 0.067 seconds
Number of clients running queries: 100
Average number of queries per client: 10
User time 0.13, System time 0.07
Maximum resident set size 26304, Integral resident set size 0
Non-physical pagefaults 514, Physical pagefaults 0, Swaps 0
Blocks in 0 out 0, Messages in 0 out 0, Signals 0
Voluntary context switches 6160, Involuntary context switches 48
[root@t3 ~]# mysqlslap -a --concurrency=50,100 --number-of-queries 1000 --iterations=5 --debug-info -uroot -p
Enter password:
Benchmark
Average number of seconds to run all queries: 0.059 seconds
Minimum number of seconds to run all queries: 0.056 seconds
Maximum number of seconds to run all queries: 0.061 seconds
Number of clients running queries: 50
Average number of queries per client: 20
Benchmark
Average number of seconds to run all queries: 0.068 seconds
Minimum number of seconds to run all queries: 0.068 seconds
Maximum number of seconds to run all queries: 0.070 seconds
Number of clients running queries: 100
Average number of queries per client: 10
User time 0.67, System time 0.35
Maximum resident set size 29952, Integral resident set size 0
Non-physical pagefaults 1268, Physical pagefaults 0, Swaps 0
Blocks in 0 out 0, Messages in 0 out 0, Signals 0
Voluntary context switches 29748, Involuntary context switches 184
测试不同引擎对比
[root@t3 ~]# mysqlslap -a --concurrency=50,100 --number-of-queries 1000 --iterations=5 --engine=myisam,innodb --debug-info -uroot -p
Enter password:
Benchmark
Running for engine myisam
Average number of seconds to run all queries: 0.059 seconds
Minimum number of seconds to run all queries: 0.058 seconds
Maximum number of seconds to run all queries: 0.061 seconds
Number of clients running queries: 50
Average number of queries per client: 20
Benchmark
Running for engine myisam
Average number of seconds to run all queries: 0.069 seconds
Minimum number of seconds to run all queries: 0.068 seconds
Maximum number of seconds to run all queries: 0.070 seconds
Number of clients running queries: 100
Average number of queries per client: 10
Benchmark
Running for engine innodb
Average number of seconds to run all queries: 2.624 seconds
Minimum number of seconds to run all queries: 2.546 seconds
Maximum number of seconds to run all queries: 2.741 seconds
Number of clients running queries: 50
Average number of queries per client: 20
Benchmark
Running for engine innodb
Average number of seconds to run all queries: 3.066 seconds
Minimum number of seconds to run all queries: 2.653 seconds
Maximum number of seconds to run all queries: 3.454 seconds
Number of clients running queries: 100
Average number of queries per client: 10
User time 1.34, System time 0.85
Maximum resident set size 30272, Integral resident set size 0
Non-physical pagefaults 2541, Physical pagefaults 0, Swaps 0
Blocks in 0 out 0, Messages in 0 out 0, Signals 0
Voluntary context switches 66867, Involuntary context switches 3948
四、测试小结
总体感觉很强大,网速也很快;
1、总体 推荐专业人士使用,第一次使用power架构,不太熟悉,测试比较片面。
2、硬件方面 cpu确实强劲,mysql运行速度很快,尤其myisam引擎表现优异。内存表现也很好,硬盘读写速度很快,超过了我所用的sata硬盘的速度。
由于时间关系,测试不太深入,请其他同学补充指正。
五、个人建议
1、管理方面建议弄一个管理虚机后台,在webshell方式管理;
2、安全方面建议增加一些简单的安全防护措施,如方ddos攻击、密码多次错误锁定、sql漏洞检测、上传漏洞检测修复等;
3、网络方面建议在现有网络情况下增加一个虚拟内网,虚拟内网支持ipsecvlan通道,在内网中可以自建更多的虚拟机和虚拟存储服务器,这样实现云数据中心的架构;我个人也相信这个是未来的方向;
4、监控报警方面建议有个人性化的界面,设置报警阈值,方便随时掌握服务器运行情况(硬件、应用和网络带宽)。
|