Skip to content

Cassandra Write Performance (2.0.7)

xamry edited this page Mar 8, 2013 · 1 revision

Recently we did performance test over current kundera release. This release enable us to fully utilize usage of secondary index support provided by cassandra. So here are details (given below) :

  • Enviornment: CentOS 6.2.

Bulk Load:

Number Of Records(Single Thread) Pelops Time (in ms) Hector Time (in ms) Kundera (in ms)
1 3 15 66
1000 167 211 1122
4000 709 931 2422
40000 6435 6429 8416
100000 14383 16417 19843
1000000 163779 162660 185139

Concurrent Load:

Number Of Threads (1 record) Pelops Time (in ms) Hector Time (in ms) Kundera (in ms)
10 8 13 24
100 54 71 244
1000 507 553 1433
10000 4449 4622 6198
40000 17129 15977 19852
50000 21835 21597 24969
100000 43643 45286 51869

Concurrent Load + Bulk Load:

Number Of Threads (1000 rec/ thread) Pelops Time (in ms) Hector Time (in ms) Kundera (in ms)
10 1813 2030 3304
100 15469 17400 20478
1000 168812 164529 196495

Source Code: Source code for producing these results is available here.

Steps to run:

  1. Download jar.
  2. java -jar {Number of records} {client type} {process mode} { number of threads} -Xmx1024m -XX:MaxPermSize=128m (e.g. java -jar /home/impadmin/workspace/kundera-cassandra-2.0.7-perf.jar 1 pelops b -Xmx1024m -XX:MaxPermSize=128m

{client type} => 1) Pelops 2) kundera 3) Hector {process mode} => 1)c(Cnoncurrent) 2) b(Bulk) 3)cb(Concurrent+Bulk)

Disclaimer:

Above mentioned results are on the specified environment and use case and might have different outcome in other cases

HOME

Clone this wiki locally