Have something interesting to share with Java Eastern Europe community?
Become a speaker now
Nikita Salnikov-Tarnovski

Nikita Salnikov-Tarnovski

Plumbr, Estonia

Co-founder of Plumbr, the memory leak detection product, where he now contributes as core developer. Besides his daily technical tasks he is an active blogger, JavaOne RockStar and frequent conference speaker (Devoxx, JavaOne Russia, 33rd Degree, TopConf, JavaDay, GeekOut, Joker, Jazoon etc).

Prior to founding Plumbr, Nikita was a Java EE developer and performance consultant in the Baltics, and has worked with tens of different Java EE applications over the years. In the last four years he specialized in troubleshooting and performance optimization.

Speaker's activity
First steps in GC tuning
May 22
45 minutes
Talk
Russian

Abstract: Correctly configured Garbage Collector is essential for application performance. Unfortunately, there are more than 200 JVM configuration parameters which influence GC in one way or another. Their very obscure names make your command line looks like a magic incantation.

The goal of this talk is to demystify Garbage Collector optimisations. I will explain the basics of Garbage Collector tuning, starting from data collection and analysis. I will show how to formulate optimisation hypotheses, choose right parameters for it and verify if they worked. You will not become a GC expert, but you will not be afraid of GC tuning anymore.

Slides:

Video:

Heap, off you go
May 23
45 minutes
Talk
Russian

Abstract: Automatic memory management and garbage collection is one of the biggest advantages of JVM compared with previous platforms. But it has a price: GC overhead and pauses. When every millisecond counts this price can be too high. You can get away and not pay it by taking responsibility back into your hands.

I will describe what does “off-heap” memory means for java application developer and how he can use it in order to speed his application up. The majority of this talk is live code demonstration. I will use different ways to implement off-heap data storage and will show how fast one and the same piece of code runs with each implementation.

Slides:

Video: