Memory Efficient Fork-based Checkpointing Mechanism for In-Memory Database Systems

Fork-based checkpointing with efficient copy-on-write management to minimize memory overhead for in-memory databases

Featured image

Venue: SAC
Slides: Google Drive

Topic: Fork-based checkpointing for in-memory databases relies on copy-on-write, but heavy write workloads during checkpointing can cause significant memory bloat. This paper proposes mechanisms to reduce that overhead.


Summary

In-memory databases use fork-based checkpointing: fork() creates a child process that writes the checkpoint while the parent continues processing transactions via copy-on-write (CoW). Problem: under high write throughput, many pages get copied → memory usage balloons, potentially exhausting system memory. This paper proposes mechanisms to make fork-based checkpointing memory-efficient for in-memory database systems.


Background

Fork-based checkpointing

Why this is a problem for in-memory databases


Key Idea


Meeting Notes

(to be filled)