Home Posts Preferences Publications People Positions

One Paper Accepted by HPDC 2026

Posted:   March 31, 2026

Status:   Completed

Tags :   News

Categories :   Publication

Were equations, pictures or diagrams not properly rendered, please refresh the page. If the problem persists, you can contact us.

Toast Lab has one technical paper accepted by the 35th ACM International Symposium on High-Performance Parallel and Distributed Computing (HPDC 2026). It is titled "Pome: Parallelizing I/Os and Computations for Efficient LSM-tree-based Data Storage". It was solely done by Toast Labmates. Mr HU Yanpeng and Ms ZHU Li have contributed equally to the paper. Mr JIA Lei is a co-author. Chundong is the corresponding author.

CPU computations and I/O operations are fundamental to data storage systems. Storage systems conduct computations with their user threads, such as sorting data for orderliness. They handle I/Os mainly through system calls (syscalls) including file write, read, and fsync, which the OS’s kernel threads perform with storage devices. Today, LSM-tree-based storage systems are widely deployed in production environments. Compaction is an essential operation that LSM-tree employs to maintain its tiered tree-like structure by re-sorting and re-storing data through computations and I/Os,respectively. In this paper, we first overhaul the procedure of a compaction. We find that computations and I/Os execute in sequential order. After re-sorting data, the user thread waits for a kernel thread to complete file write and fsync I/Os. These costly synchronous I/Os create a severely long critical path that affects the performance of LSM-tree. To address this issue, we propose parallelizing I/Os and computations for efficient LSM-tree-based data storage (Pome). Pome decouples computations from I/Os within each compaction by referring to its new protocol that moves I/O operations out of the critical path. To this end, it leverages the io_uring to perform asynchronous I/Os. Furthermore, regarding the potential I/O congestion caused by accelerated compactions, Pome incorporates an adaptive I/O rate limiter to achieve smooth execution. We prototype Pome on top of RocksDB. Experimental results demonstrate that Pome significantly improves the performance of RocksDB and outperforms several state-of-the-art LSM-tree variants.

This year, HPDC received 281 submissions and accepted 41 papers for the technical program and 28 for poster presentations, for an acceptance rate of just over 14% for technical papers. It will convene in Cleverland, Ohio, USA during 13-16, 2026.