# Gang-GC: Locality-aware Parallel Data Placement Optimizations for Key-Value Storages

## TL;DR

This paper introduces Gang-GC, a locality-aware extension to parallel garbage collection in JVMs, improving data access speed and reducing page faults in key-value store applications without requiring application profiling or special data structures.

## Contribution

Gang-GC extends heap layouts and GC algorithms to enhance locality among related objects, improving performance in key-value stores without application profiling.

## Key findings

- 6% improvement in read and update operation times
- 12.4% reduction in page faults
- Negligible overhead in object allocation

## Abstract

Many cloud applications rely on fast and non-relational storage to aid in the processing of large amounts of data. Managed runtimes are now widely used to support the execution of several storage solutions of the NoSQL movement, particularly when dealing with big data key-value store-driven applications. The benefits of these runtimes can however be limited by modern parallel throughput-oriented GC algorithms, where related objects have the potential to be dispersed in memory, either in the same or different generations. In the long run this causes more page faults and degradation of locality on system-level memory caches. We propose, Gang-CG, an extension to modern heap layouts and to a parallel GC algorithm to promote locality between groups of related objects. This is done without extensive profiling of the applications and in a way that is transparent to the programmer, without the need to use specialized data structures. The heap layout and algorithmic extensions were implemented over the Parallel Scavenge garbage collector of the HotSpot JVM\@. Using microbenchmarks that capture the architecture of several key-value stores databases, we show negligible overhead in frequent operations such as the allocation of new objects and improvements to the access speed of data, supported by lower misses in system-level memory caches. Overall, we show a 6\% improvement in the average time of read and update operations and an average decrease of 12.4\% in page faults.