Saturday, July 08, 2006

More info on iSCSI benchmark

Looks like a lot of interest was generated from the results of RightMedia's iSCSI benchmarking. Peter Zaitsev is curious about the environment of the benchmark. So here's all the gory details
MySQL version : 5.0.21
System : GenToo Linux, kernel version 2.6.16-gentoo-r2-1
Hardware : 4GB memory, 2 Dual Core AMD Opteron 275 chips
MySQL configuration file (abridged version):

[client]
port = 3306
socket = /var/run/mysqld/mysqld.sock

[mysqld]
basedir = /usr
datadir = /var/lib/mysql/data
port = 3306
pid-file = /var/run/mysqld/mysqld.pid
log-error = /var/log/mysql/mysqld.err
socket = /var/run/mysqld/mysqld.sock
user = mysql
max_connections = 1024
table_cache = 128
thread_case_size = 32
sort_buffer_size = 64M
innodb_additional_mem_pool_size = 20M
innodb_buffer_pool_size = 3000M
innodb_data_home_dir = /var/lib/mysql/inno
innodb_log_group_home_dir = /var/lib/mysql/log/inno
innodb_log_file_size = 2000M
innodb_log_files_in_group = 2
innodb_log_buffer_size = 8M
innodb_flush_trx_at_commit = 2
innodb_autoincrement_size = 32M
innodb_status_file
innodb_thread_concurrency = 4
innodb_flush_method = O_DIRECT
innodb_support_xa = 0

sysbench v0.4.3
Command used to prepare 100,000,000 record sbtest database:
sysbench --num-threads=1 --test=oltp --mysql-user=root --oltp-table-size=100000000 --mysql-table-type=innodb prepare

For each SysBench run:
sysbench --num-threads=# --test=oltp --mysql-user=root --oltp-table-size=100000000 --mysql-table-type=innodb --max-requests=10000 --oltp-test-mode=complex run

Before each SysBench run, the kernel pagecache was also cleaned out:
sync ; echo 1 > /proc/sys/vm/drop_caches

This was done to provide a fair & clean test. For each batch of runs by thread, 4 runs were made,with the best TPS used as the final number for the batch.

I am pretty happy with the results. I plan on putting for iSCSI benchmark results soon, as I think a lot of folks would be interested in the results, especially if the iSCSI device scales well.

6 comments:

Peter Zaitsev said...

Thank you for clarifying results.
I still would like to see some VMSTAT printout however.

Looking at parameters you used you used right database but you used default access distribution which is very skewed - in majority of cases rows from very small primary key range will be accessed this is why performance is that high and difference is that insignificant.

It does not mean iSCSI will do bad for you in real life, just make sure you tested with load similar to what you will have in real life.

Speaking about clearing caches - this is one of the ways you can do it but this makes benchmark results unstable in the sense - 1min benchmark will return different results from 15min benchmark. I usually prefer "hot" benchmark with warmup done and only "steady state" measured. This is pretty much as TPC-C or SpecJAppServer run as well.

Happy tuning. If you publish any more benchmarks let me know. I'm always hungry for new results.

Jeremy Cole said...

how long did it take to run the "prepare" stage for each?

Partha Dutta said...

Unfortunately I had to give up my testing boxes to put into production use. On the iSCSI systen, a "prepare" run of 10,000,000 records took approximately 8 minutes. This is with a 3GB InnoDB buffer pool, and innodb_flush_method=O_DIRECT

Anonymous said...

看房子,買房子,建商自售,自售,台北新成屋,台北豪宅,新成屋,豪宅,美髮儀器,美髮,儀器,髮型,EMBA,MBA,學位,EMBA,專業認證,認證課程,博士學位,DBA,PHD,在職進修,碩士學位,推廣教育,DBA,進修課程,碩士學位,網路廣告,關鍵字廣告,關鍵字,廣告,課程介紹,學分班,文憑,牛樟芝,段木,牛樟菇,日式料理, 台北居酒屋,燒肉,結婚,婚宴場地,推車飲茶,港式點心,尾牙春酒,台北住宿,國內訂房,台北HOTEL,台北婚宴,飯店優惠,台北結婚,婚宴場地,推車飲茶,港式點心,尾牙春酒,住宿,訂房,HOTEL,飯店,造型系列,學位,牛樟芝,腦磷脂,磷脂絲胺酸,SEO,婚宴,捷運,學區,美髮,儀器,髮型,牛樟芝,腦磷脂,磷脂絲胺酸,看房子,買房子,建商自售,自售,房子,捷運,學區,台北新成屋,台北豪宅,新成屋,豪宅,學位,碩士學位,進修,在職進修, 課程,教育,學位,證照,mba,文憑,學分班,網路廣告,關鍵字廣告,關鍵字,SEO,关键词,网络广告,关键词广告,SEO,关键词,网络广告,关键词广告,SEO,台北住宿,國內訂房,台北HOTEL,台北婚宴,飯店優惠,住宿,訂房,HOTEL,飯店,婚宴,台北住宿,國內訂房,台北HOTEL,台北婚宴,飯店優惠,住宿,訂房,HOTEL,飯店,婚宴,台北住宿,國內訂房,台北HOTEL,台北婚宴,飯店優惠,住宿,訂房,HOTEL,飯店,婚宴,結婚,婚宴場地,推車飲茶,港式點心,尾牙春酒,台北結婚,婚宴場地,推車飲茶,港式點心,尾牙春酒,結婚,婚宴場地,推車飲茶,港式點心,尾牙春酒,台北結婚,婚宴場地,推車飲茶,港式點心,尾牙春酒,結婚,婚宴場地,推車飲茶,港式點心,尾牙春酒,台北結婚,婚宴場地,推車飲茶,港式點心,尾牙春酒,居酒屋,燒烤,美髮,儀器,髮型,美髮,儀器,髮型,美髮,儀器,髮型,美髮,儀器,髮型,小套房,小套房,進修,在職進修,留學,證照,MBA,EMBA,留學,MBA,EMBA,留學,進修,在職進修,牛樟芝,段木,牛樟菇,住宿,民宿,飯宿,旅遊,住宿,民宿,飯宿,旅遊,住宿,民宿,飯宿,旅遊,住宿,民宿,飯宿,旅遊,住宿,民宿,飯宿,旅遊,住宿,民宿,飯宿,旅遊,住宿,民宿,飯宿,旅遊,美容,美髮,整形,造型,美容,美髮,整形,造型,美容,美髮,整形,造型,美容,美髮,整形,造型,美容,美髮,整形,造型,美容,美髮,整形,造型,美容,美髮,整形,造型,設計,室內設計,裝潢,房地產,設計,室內設計,裝潢,房地產,設計,室內設計,裝潢,房地產,設計,室內設計,裝潢,房地產,設計,室內設計,裝潢,房地產,設計,室內設計,裝潢,房地產,設計,室內設計,裝潢,房地產,設計,室內設計,裝潢,房地產,進修,在職進修,MBA,EMBA,進修,在職進修,MBA,EMBA,進修,在職進修,MBA,EMBA,進修,在職進修,MBA,EMBA,進修,在職進修,MBA,EMBA,進修,在職進修,MBA,EMBA,進修,在職進修,MBA,EMBA,住宿,民宿,飯店,旅遊,美容,美髮,整形,造型,設計,室內設計,裝潢,房地產,進修,在職進修,MBA,EMBA,關鍵字排名,網路行銷,关键词排名,网络营销,網路行銷,關鍵字排名,关键词排名,网络营销,羅志祥,周杰倫,五月天,蔡依林,林志玲,羅志祥,周杰倫,五月天,蔡依林,林志玲,PMP,在職專班,研究所在職專班,碩士在職專班,PMP,證照,在職專班,研究所在職專班,碩士在職專班

huangtiao said...

Runes of Magic is a Massively Multiplayer Online Role playing Game without monthly costs all kinds of Rom Gold and free to download. The premise is that you have played this game and understand the necessary Runes of Magic Gold in the Runes of Magic game. Perhaps the first monster I saw taken down in the demo can gain those varieties of Runes of Magic money. Characters choose one class; the important one is that we have to buy Rom Gold first. You can switch freely between the two classes, but you can only use primary abilities from your primary class, along with cheap Runes of Magic Gold, secondary abilities from your secondary class.
It is easy to control and pick up more all kinds of latale online gold and weapons or equipment which were dropped by those monsters. You will experience a wide range of attack and strikingly gorgeous combat effects with your own latale gold. You can also freely customize your character by equipping cool armors and weapons which those you must first have need to buy latale online gold and then you can have them. There is no doubt La Tale and its very cheap latale gold will attract female gamer and as well as young gamer eyeballs. You see my problem is that I do not what class is good or fun to play without any more necessary latale money.

Anonymous said...

So to have a good attack, you should level your skill in cabal online alz. Also, get the highest level jewelry that you can equip in thecabal alz. This step depends on your skill rank, for lower magic ranks get 4 fast notice arrow skills and your cabal gold. Of course, if you have more cabal money, you can get more in the game. By the way, you can buy pet from other players in buy cabal alz.
Many people play the game maybe feel that they were too poor, some of my friend often ask, why you play games so there is ghost gen.
Say good pull so much I do not pull hedge with the US to talk about the game experience to make ghost online gen. I want to see if there is nothing worth ghost gold things and a very low price. The most important thing is you have ghost online gold. Also we have a good time happy in the game and hope earned more ghost money in the game.