发表机构
Spiral; University of Pisa; CWI（Spiral; 比萨大学; 荷兰数学与计算机科学研究中心）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
FastPair是一种GPU解码器，通过优化字典解码中的查找重组和连续输出写入，在B300上实现比DE快2.4-4.2倍的解码速度，峰值达1.6 TB/s。
AI 中文摘要
现代数据系统对静态数据进行压缩，仅在需要时解压以保持互连带宽。这种设计在基于GPU的计算平台上往往效率低下，因为许多传统压缩技术存在串行数据依赖，限制了GPU的并行性，导致资源闲置。最近的NVIDIA GPU通过解压引擎（DE）解决了这一解码缺陷，该引擎是一种片上固定功能解压加速器，支持Deflate、LZ4和Snappy等通用压缩格式。近期工作提出了字符串编解码器，用来自小型训练字典的固定宽度代码替换频繁出现的子串，使每个代码的查找独立。虽然这些查找可以并行执行，但产生的分散读取和短输出写入仍与GPU硬件不完全匹配，后者更擅长处理连续内存访问。我们提出了FastPair，一种GPU解码器，通过重新组织查找并组装解码后的子串以实现连续输出写入，优化了现有的字典解码过程。在B300上，FastPair解码十个真实世界列的速度比DE快2.4到4.2倍，达到最高1.6 TB/s。
英文摘要
Modern data systems compress data at rest and decompress it only when needed to preserve interconnect bandwidth. This design is often inefficient on GPU-based compute platforms because many conventional compression techniques exhibit serial data dependencies that limit GPU parallelism, leaving resources idle. Recent NVIDIA GPUs address this decoding deficiency through the Decompression Engine (DE), an on-die, fixed-function decompression accelerator for general-purpose compression formats such as Deflate, LZ4, and Snappy. Recent work has proposed string codecs that replace frequent substrings with fixed-width codes from a small, trained dictionary, making each code's lookup independent. While these lookups can run in parallel, the resulting scattered reads and short output writes still do not align well with GPU hardware, which handles contiguous memory accesses more efficiently. We present FastPair, a GPU decoder that optimizes the existing dictionary decoding process by reorganizing lookups and assembling decoded substrings for contiguous output writes. On a B300, FastPair decodes ten real-world columns 2.4 to 4.2x faster than the DE, reaching up to 1.6 TB/s.
