DAXFS: A Lock-Free Shared Filesystem for CXL Disaggregated Memory

2026-04-02

AI Summary

Existing file systems struggle to support lock-free, multi-host concurrent access in CXL shared memory environments. This work proposes the first decentralized Linux file system leveraging CXL 3.0 shared memory, relying solely on cmpxchg atomic operations for cross-host coordination. It introduces a CAS-based hash overlay mechanism to enable concurrent writes and designs an MH-clock algorithm for managing the shared page cache. Integrated with DAX for direct storage access, the system achieves over 99% CAS success rates under high contention without lost updates. On a single-node DRAM backend, it delivers 2.68× higher random write throughput and 1.18× higher random read throughput compared to tmpfs, and demonstrates that GPUs can efficiently access the shared page cache via PCIe 5.0 bandwidth.

Abstract

CXL (Compute Express Link) enables multiple hosts to share byte-addressable memory with hardware cache coherence, but no existing filesystem exploits this for lock-free multi-host coordination. We present DaxFS, a Linux filesystem for CXL shared memory that uses cmpxchg atomic operations, which CXL makes coherent across host boundaries, as its sole coordination primitive. A CAS-based hash overlay enables lock-free concurrent writes from multiple hosts without any centralized coordinator. A cooperative shared page cache with a novel multi-host clock eviction algorithm (MH-clock) provides demand-paged caching in shared DAX memory, with fully decentralized victim selection via cmpxchg. We validate multi-host correctness using QEMU-emulated CXL 3.0, where two virtual hosts share a memory region with TCP-forwarded atomics. Under cross-host contention, DaxFS maintains >99% CAS accuracy with no lost updates. On single-host DRAM-backed DAX, DaxFS exceeds tmpfs throughput across all write workloads, achieving up to 2.68x higher random write throughput with 4 threads and 1.18x higher random read throughput at 64 KB. Preliminary GPU microbenchmarks show that the cmpxchg-based design extends to GPU threads performing page cache operations at PCIe 5.0 bandwidth limits.

Problem

• CXL
• disaggregated memory
• lock-free
• filesystem
• multi-host coordination

Innovation

• CXL
• lock-free
• cmpxchg
• shared memory filesystem
• multi-host coordination