MARS: Multi-stage Accelerated Read Stack for Large-buffer Buffered Reads
For data-intensive applications relying on large-buffer reads, MARS significantly improves I/O performance by rearchitecting the kernel's page-cache handling.
MARS redesigns the Linux buffered read path for large-buffer reads, achieving up to 6.56x bandwidth improvement over Linux in fio benchmarks and 36.87 GiB/s on RAID0 NVMe SSDs, while accelerating DuckDB queries by 1.80-2.15x and model loading by 3.17-3.61x.
Large-buffer reads increasingly connect data-intensive applications to high-speed storage. They amortize system-call overhead and create a larger in-kernel window for organizing page-cache work and submitting I/O. However, Linux buffered read primarily exploits only the former benefit. Within a large read, its conventional interleaved path repeatedly switches among fine-grained page-cache operations, amplifying metadata and serial orchestration overheads and failing to consistently expose enough in-flight requests to modern parallel SSDs. We present MARS, a multi-stage accelerated read stack for synchronous large-buffer buffered reads. MARS treats each large-range read as one unit of work and stages page-cache operations by data structure and dependency. During I/O waits, it handles user-buffer page faults and performs reorderable data copies early. Opportunistic kernel workers then copy remaining data in parallel and, when the backend provides sufficient parallelism, optionally submit I/O in parallel. We implement MARS in Linux 6.6.58. For MiB-scale fio reads, MARS improves bandwidth by up to 6.56 times over Linux. On five NVMe SSDs in RAID0, it reaches 36.87 GiB/s for 128 MiB random reads, 4.44 times Linux. MARS also accelerates DuckDB/Parquet queries by 1.80--2.15 times and ExecuTorch model loading by 3.17--3.61 times.