The Cassandra database is a widely deployed Java infrastructure piece, and a critical part of many services. A core part of it's work is deduplicating and deleting data through periodic compactions. In this talk I will discuss a recent patch in which I re-implemented this process and deliver a large performance improvement. This is an exploration of the delivery of a meaningful performance improvement to a long standing, industry standard, OSS project.
Key Takeaways:
Audience:
Key Takeaways:
- Picking the right target
- Setting meaningful goals and metrics
- The contribution process
- Re-occurring optimisation techniques
Audience:
- Cassandra enthusiasts
- Java performance enthusiasts
- Intermediate/Senior Java eng
Nitsan Wakart
Apple
A blogger and a coder with a pedantic passion for performance. My work has spanned army intelligence systems, dot.com era startups, financial institutions and innovative product companies. A JVM Engineer for Apple by day, main contributor to lock free data structures library JCTools by night.