Report this

What is the reason for this report?

redis-benchmark comparison

Posted on November 9, 2013

Hey guys, I was just curious on what other people’s redis-benchmark results here on DO using. I was using the usual: “redis-benchmark -q -n 100000” command in both my 512mb droplet and 2gb droplet and was surprised that the smaller droplet had better results almost all the time (both are empty).

2GB 2VCPU: [root@redis ~]# redis-benchmark -q -n 100000 PING_INLINE: 34977.27 requests per second PING_BULK: 27570.99 requests per second SET: 24213.07 requests per second GET: 22517.45 requests per second INCR: 27107.62 requests per second LPUSH: 37495.31 requests per second LPOP: 28604.12 requests per second SADD: 24283.63 requests per second SPOP: 23413.72 requests per second LPUSH (needed to benchmark LRANGE): 21953.90 requests per second LRANGE_100 (first 100 elements): 15900.78 requests per second LRANGE_300 (first 300 elements): 7773.63 requests per second LRANGE_500 (first 450 elements): 5799.79 requests per second LRANGE_600 (first 600 elements): 4857.91 requests per second MSET (10 keys): 22158.21 requests per second

512MB 1VCPU: [root@ramonapps ~]# redis-benchmark -q -n 100000 PING_INLINE: 35971.22 requests per second PING_BULK: 36049.03 requests per second SET: 36114.12 requests per second GET: 36913.99 requests per second INCR: 36589.83 requests per second LPUSH: 31756.11 requests per second LPOP: 32819.17 requests per second SADD: 30376.67 requests per second SPOP: 29664.79 requests per second LPUSH (needed to benchmark LRANGE): 30553.01 requests per second LRANGE_100 (first 100 elements): 15398.83 requests per second LRANGE_300 (first 300 elements): 6976.91 requests per second LRANGE_500 (first 450 elements): 5290.45 requests per second LRANGE_600 (first 600 elements): 4490.14 requests per second MSET (10 keys): 20012.01 requests per second

I know redis has latency problems with Xen, but I didn’t expect the one with 2 cpus get beaten by the one with 1 cpu.



This textbox defaults to using Markdown to format your answer.

You can type !ref in this text area to quickly search our full set of tutorials, documentation & marketplace offerings and insert the link!

These answers are provided by our Community. If you find them useful, show some love by clicking the heart. If you run into issues leave a comment, or add your own answer to help others.

edit: I just realized DO uses KVM instead of Xen.

The above stats are very poor atleast 1/3rd of what you can expect on a basic box. Don’t know why it’s soo.

Yes and Redis is mono-threaded. It will not use both cores. <br>The only advantage is to have more memory

The developer cloud

Scale up as you grow — whether you're running one virtual machine or ten thousand.

Get started for free

Sign up and get $200 in credit for your first 60 days with DigitalOcean.*

*This promotional offer applies to new accounts only.