The Cost of Garbage Collection for State Machine Replication

TL;DR

This paper systematically compares the performance and resource costs of implementing a state machine replication system in languages with garbage collection versus manual memory management, revealing implications for cloud system design.

Contribution

It provides the first systematic study of GC overhead in SMR-based cloud systems by implementing and benchmarking a MultiPaxos key-value store in C++, Java, Rust, and Go.

Findings

GC overhead impacts latency and resource usage

Manual memory management can reduce cloud costs

Performance varies significantly across languages

Abstract

State Machine Replication (SMR) protocols form the backbone of many distributed systems. Enterprises and startups increasingly build their distributed systems on the cloud due to its many advantages, such as scalability and cost-effectiveness. One of the first technical questions companies face when building a system on the cloud is which programming language to use. Among many factors that go into this decision is whether to use a language with garbage collection (GC), such as Java or Go, or a language with manual memory management, such as C++ or Rust. Today, companies predominantly prefer languages with GC, like Go, Kotlin, or even Python, due to ease of development; however, there is no free lunch: GC costs resources (memory and CPU) and performance (long tail latencies due to GC pauses). While there have been anecdotal reports of reduced cloud cost and improved tail latencies when switching from a language with GC to a language with manual memory management, so far, there has not been a systematic study of the GC overhead of running an SMR-based cloud system. This paper studies the overhead of running an SMR-based cloud system written in a language with GC. To this end, we design from scratch a canonical SMR system -- a MultiPaxos-based replicated in-memory key-value store -- and we implement it in C++, Java, Rust, and Go. We compare the performance and resource usage of these implementations when running on the cloud under different workloads and resource constraints and report our results. Our findings have implications for the design of cloud systems.