Great Product

• By Brian
• on 02/15/2018

We were very glad to find Heimdall Data and eventually implement their product. Their platform handles the caching layer for database queries, among many other things. The only comparable solution we came across in the market is Pgpool, but Heimdall Data has proven to be in a class of its own.

Pgpool handles connection pooling, read/write splitting, and manual caching, However, it is all manual and requires code changes. Pgpool also does not support SELECT FOR UPDATE and several other options we found readily available through Heimdall Data.

We first tested Pgpool and very quickly realized that it would require extensive work to update, maintain and scale. When we began testing with Heimdall Data we were able to see huge improvements in load testing very quickly due to its automation. We were able to increase our test throughput about 5-7x what original performance tests had shown. We continued to test and learned all about the rules engine, explain plans, and analytics which allowed us to quickly identify several places for optimization. We could easily see query response sizes, response times, and query count, among others.

Heimdall data has allowed us to see into the database queries and transaction times on both the application and Amazon Aurora. The level of monitoring and analytics that Heimdall Data provides allowed us to find and identify issues that only a seasoned DBA would be able to gather previously.

We deployed Heimdall Data to our production environment and have been seeing between 60-70% cache hit rate overall. Immediately we noticed an improvement in response times and the benefit of our application and database load being reduced greatly. We have been running Heimdall Data in production for over 2 months and have had great results. This has been an invaluable tool for helping us scale and allowing us to keep resources low.

Overall, We have been very satisfied with Heimdall Data and can highly recommend it.