GBDI: Going Beyond Base-Delta-Immediate Compression with Global Bases
Alexandra Angerd, Angelos Arelakis, Vasilis Spiliopoulos, Erik Sintorn, Per Stenström
摘要
Memory bandwidth is limiting performance for many emerging applications. While compression techniques can unlock a higher memory bandwidth, prior art offers only modestly better bandwidth. This paper contributes with a new compression method – Global Base Delta Immediate compression (GBDI) – that offers substantially higher memory bandwidth by, unlike prior art, selecting base values across memory blocks. GBDI uses a novel clustering algorithm through data analysis in the background. The presented accelerator infrastructure offers low area overhead and latency. This paper shows that GBDI offers a compression ratio of 2.3×, and yields 1.5× higher bandwidth and 1.1× higher performance compared with a baseline without compression support, on average, for SPEC2017 benchmarks requiring medium to high memory bandwidth.
问问这篇 Paper
问问你的智能体。
Lune 读过与它相关的顶会 Paper,每个回答都会注明依据哪几篇。
引用它的顶会 Paper4
- SAGe: A Lightweight Algorithm-Architecture Co-Design for Mitigating the Data Preparation Bottleneck in Large-Scale Genome Sequence AnalysisNika Mansouri-Ghiasi, Talu Güloglu, Harun Mustafa, Can Firtina 等HPCA 2026 · 被引用 3 次
- Memory Allocation Under Hardware CompressionMuhammad Laghari, Yuqing Liu, Gagandeep Panwar, David Bears 等MICRO 2024 · 被引用 2 次
- The XOR Cache: A Catalyst for CompressionZhewen Pan, Joshua San MiguelISCA 2025 · 被引用 1 次
- MedFS: Pursuing Low Update Overhead via Metadata-Enabled Delta Compression for Log-structured File System on Mobile DeviceChao Wu, Cheng Ji, Li-Pin Chang, Zongwei Zhu 等FAST 2025
相关 Paper
- Exploiting Inter-block Entropy to Enhance the Compressibility of Blocks with Diverse DataJinkwon Kim, Mincheol Kang, Jeongkyu Hong, Soontae KimHPCA 2022 · 被引用 5 次
- Random-Access Hardware Sequence CompressionNolan Chu, Yoon Lee, Gagandeep Panwar, Xun Steve JianISCA 2026
- Buddy Compression: Enabling Larger Memory for Deep Learning and HPC Workloads on GPUsEsha Choukse, Michael B. Sullivan, Mike O'Connor, Mattan Erez 等ISCA 2020 · 被引用 58 次
- Baryon: Efficient Hybrid Memory Management with Compression and Sub-BlockingYiwei Li, Mingyu GaoHPCA 2023 · 被引用 14 次
- Thesaurus: Efficient Cache Compression via Dynamic ClusteringAmin Ghasemazar, Prashant J. Nair, Mieszko LisASPLOS 2020 · 被引用 22 次
