# New high performance plan - Loud Lion

By Lovisa Johansson. Published 5 November 2015.

We are happy to announce a new high performance plan available for the public, Loud Lion. Loud Lion is for customers with very high message rate requirements!

![New plan Loud Lion](/img/plans/lion_256.png "new plan loud lion")

If you create this plan in [AWS](https://aws.amazon.com) it's two c3.4xlarge instances with 30GB memory and 16 high performance cores underneath. In [GCE](https://cloud.google.com/products/compute-engine/) it's two n1-highcpu-32 instances, with 28.8GB memory and 32 cores(!).

### Benchmark testing

The [RabbitMQ PerfTest tool](https://www.rabbitmq.com/java-tools.html) is part of the RabbitMQ Java client and it’s the tool we use for benchmark testing of the plans. The library is used to generate load on the cluster. The benchmark tests for Loud Lion on AWS for two nodes (in the same region) shows that you can reach about 40.000 messages per second throughput with one queue, one publisher and one customer. Now keep in mind that queues in RabbitMQ are limited to one core - when we were using more then one queue and more producers in the benchmark testing **we could easily get 300.000 messages per second throughput.**

#### Test details

`rabbitmq-java-client-bin-3.5.4/runjava.sh com.rabbitmq.examples.PerfTest --uri $uri --consumers 12 --producers 4 --autoack`

![loud lion](/img/blog/loud_lion_benchmark.png "loud lion")

time: 24.000s, sent: 34589 msg/s, received: 333531 msg/s  
time: 25.000s, sent: 15570 msg/s, received: 326635 msg/s  
time: 26.000s, sent: 33261 msg/s, received: 334142 msg/s  
time: 27.000s, sent: 37146 msg/s, received: 336071 msg/s  
time: 28.000s, sent: 27254 msg/s, received: 330693 msg/s

Give it a try and do some benchmark testing! As usual, all plans are billed by the second. Please email us at [contact@cloudamqp.com](mailto:contact@cloudamqp.com) if you have any suggestions or feedback.
