Every so often we come across a use case that makes every hour of work put into our Full Metal Cloud worth it many times over. Today we’re in the happy position of sharing one of those use cases with you. AlignAlytics ran Elasticsearch queries on 10 million documents (approx. 4 GB of compressed data) and consistently saw with Bigstep a 100-200% performance improvement over their existing dedicated servers. The specs on their machines were quite similar to those of our Full Metal Compute Instances, which makes this one of the closest “apples to apples” comparisons we’ve done.
Many IT directors are failing to get the optimum performance from their infrastructure. So we’ve been conducting a number of benchmarking studies to see to how this can be improved.
Our product manager Alex Bordei presented the findings on an O’Reilly webcast which was hosted by O’Reilly’s Chief Data Scientist and Director of Content Strategy for Data, Ben Lorica, a true big data luminary.
Alex Bordei (@alexandrubordei), our Product Manager has benchmarked Elasticsearch on a wide range of Full Metal Clusters, in order to find the infrastructure setup that provides the best price/performance.
We went from clusters of just 2 very powerful Full Metal Compute Instances (192 GB RAM) to clusters with 14 small instances (8 GB RAM) – and tested pretty much everything in between.
This presentation summarizes our most interesting findings. It really might help you make better decisions about the architecture of your Elasticsearch cluster, as it shows how Elasticsearch scales vertically and horizontally and when it might be worth it to do either.
If you have any questions, let us know and we’ll do our best to answer.