WIO: Upload-Enabled Computational Storage on CXL SSDs

TL;DR

WIO introduces a reversible, dynamic computation migration system on CXL SSDs using WebAssembly, improving throughput and latency by adapting to thermal and power constraints without modifying applications.

Contribution

It presents extsc{WIO}, a system enabling reversible computation migration on CXL SSDs with WebAssembly, addressing thermal and power challenges dynamically.

Findings

Up to 2× throughput improvement

3.75× reduction in write latency

Transforms thermal cliffs into elastic trade-offs

Abstract

The widening gap between processor speed and storage latency has made data movement a dominant bottleneck in modern systems. Two lines of storage-layer innovation attempted to close this gap: persistent memory shortened the latency hierarchy, while computational storage devices pushed processing toward the data. Neither has displaced conventional NVMe SSDs at scale, largely due to programming complexity, ecosystem fragmentation, and thermal/power cliffs under sustained load. We argue that storage-side compute should be \emph{reversible}: computation should migrate dynamically between host and device based on runtime conditions. We present \sys, which realizes this principle on CXL SSDs by decomposing I/O-path logic into migratable \emph{storage actors} compiled to WebAssembly. Actors share state through coherent CXL.mem regions; an agility-aware scheduler migrates them via a zero-copy drain-and-switch protocol when thermal or power constraints arise. Our evaluation on an FPGA-based CXL SSD prototype and two production CSDs shows that \sys turns hard thermal cliffs into elastic trade-offs, achieving up to 2$\times$ throughput improvement and 3.75$\times$ write latency reduction without application modification.