Automatic Query Cache

What: Our proxy caching module provides a transparent look-aside cache solution. We automatically cache SQL results offloading your database.

How: You choose the cache, and Heimdall provides the cache logic determining what should be cached and invalidated.

Benefit: Save database costs by offloading SQL traffic. Improve response times. Save months of software development and management per year. There is no complex coding or configuration required. The Heimdall proxy provides up to 20x in increased performance. 

  • For the cache storage, choose 1) In-memory in the application instance, 2) Grid-cache or 3) a combination of both. Below are supported grid-caches:
    • Redis
    • Amazon Elasticache
    • Hazelcast
    • Oracle Coherence
    • Pivotal GemFire / Geode
    • GridGain / Apache Ignite
  • Automated cache invalidation
  • Automated cache refresh upon invalidation
  • Cache memory and queries cache rules are configurable at run time

Question & Answer:

Q: What about caching stored procedures?

A: We can cache those too! Heimdall detects when unknown stored procedures were accessed. The proxy will pull the text of the stored procedure, parse out if it uses return values or not, and cache this information.

Q: Can you cache within a transaction?

A: Yes. Our analytics give you visibility on whether it is safe to cache within a transaction. Additionally, we have “transaction suppression”: Although the application sends a transaction, we will delay the transaction to the database and cache until the proxy sees the DML. This allows the proxy to cache the read queries safely. For applications that are heavily in transactions, we now have an opportunity to automate caching. The Odoo E-commerce application is one such example.

Download a free trial.