# Benchmarking MySQL Cluster (NDB 7.1.17) using TPC-E benchmarktool

**URL:** <https://forums.percona.com/t/benchmarking-mysql-cluster-ndb-7-1-17-using-tpc-e-benchmarktool/2381>\
**Category:** Other MySQL® Questions\
**Created:** [December 21, 2011, 6:49am UTC](https://forums.percona.com/t/benchmarking-mysql-cluster-ndb-7-1-17-using-tpc-e-benchmarktool/2381 "2011-12-21T06:49:54Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![swetha](https://avatars.discourse-cdn.com/v4/letter/s/ebca7d/32.png) [@swetha](https://forums.percona.com/u/swetha)\
**Post date:** [December 21, 2011, 6:49am UTC](https://forums.percona.com/t/benchmarking-mysql-cluster-ndb-7-1-17-using-tpc-e-benchmarktool/2381/1 "2011-12-21T06:49:54Z")

</div>

Hi,

I have a MySQL Cluster(NDB 7.1.17) setup. The storage engine i used was NDB Cluster.I tried benchmarking this using the TPC-E tool. The throughput that i got was only around 1 tpsE.

For the same setup i created a database with InnoDB as the storage engine. When i ran the test using InnoDB as the storage engine, i received more throughput (240tpsE).

I thought the issue was with the NDB. Can u help me in narrowing down the issue.

Thanks in Advance

---

<div class="post-metadata">

**Author:** ![system](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/system/32/9292_2.png) [@system](https://forums.percona.com/u/system)\
**Post date:** [December 26, 2011, 11:13am UTC](https://forums.percona.com/t/benchmarking-mysql-cluster-ndb-7-1-17-using-tpc-e-benchmarktool/2381/2 "2011-12-26T11:13:41Z")

</div>

MySQL Cluster and NDB Storage Engine has a lot of environment variables which you must take with in order to get your MySQL Cluster up & running with acceptable performance. Unfortunately, NDB Storage engine is not like InnoDB that has _good_ start parametrization to compare it with InnoDB.

You should check it out this links:

=\> [http://www.severalnines.com/downloads/presentations/MySQL\_UC](http://www.severalnines.com/downloads/presentations/MySQL_UC) 2010\_Cluster\_PT.pdf  
=\> [URL][http://www.mysql.com/why-mysql/performance/index.html[/URL]](http://www.mysql.com/why-mysql/performance/index.html%5B/URL%5D)

Another thing: before to start over taking with MySQL Cluster again, you must understand that everything (indexes and data) is intended to stay into memory and to have better notion of how much memory you must dedicate to indexes and data in data nodes machines, run ndb\_size.pl.

[http://dev.mysql.com/doc/refman/5.1/en/mysql-cluster-program](http://dev.mysql.com/doc/refman/5.1/en/mysql-cluster-program) s-ndb-size-pl.html

I am looking forward to know more about your tests.

🙂