COAXIAL: A CXL-Centric Memory System for Scalable Servers
Albert Cho, Anish Saxena, Moinuddin Qureshi, Alexandros Daglis
摘要
The memory system is a major performance determinant for server processors. Ever-growing core counts and datasets demand higher memory bandwidth and capacity. DDR—the dominant processor interface to memory—requires a large number of on-chip pins, which is a scarce resource, thus limiting the processor’s memory bandwidth. With limited bandwidth, multiple concurrent memory requests experience significant queuing delays that often overshadow DRAM’s service time and degrade performance. We present CoaXial, a memory system design for throughput-oriented manycore servers that replaces all of the processor’s DDR interfaces with the pin-efficient CXL interface, which offers higher bandwidth per pin. While such replacement incurs a considerable latency overhead, we demonstrate that, for many workloads, and with careful integration, CXL’s higher bandwidth more than offsets its latency premium. Our evaluation shows that CoaXial improves the performance of manycore throughput-oriented servers by on average and by up to .
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper3
- CXL Memory Performance for In-Memory Data ProcessingMarcel Weisgut, Daniel Ritter, Pinar Tözün, Lawrence Benson 等VLDB 2025 · 被引用 7 次
- Break On Through to the Other Side: Pooling Memory Elastically with RamRyderYanbo Zhou, Erci Xu, Dongjoo Seo, Adam Manzanares 等OSDI 2026
- MAC: Metadata Acceleration for Sustainable Performance in Big-Data Systems with CXL DRAMDusol Lee, Yan Sun, Houxiang Ji, Vinit Gupta 等OSDI 2026
它引用的顶会 Paper4
- Pond: CXL-Based Memory Pooling Systems for Cloud PlatformsHuaicheng Li, Daniel S. Berger, Lisa Hsu, Daniel Ernst 等ASPLOS 2023 · 被引用 328 次
- TPP: Transparent Page Placement for CXL-Enabled Tiered-MemoryHasan Al Maruf, Hao Wang, Abhishek Dhanotia, Johannes Weiner 等ASPLOS 2023 · 被引用 255 次
- Demystifying CXL Memory with Genuine CXL-Ready Systems and DevicesYan Sun, Yifan Yuan, Zeduo Yu, Reese Kuper 等MICRO 2023 · 被引用 133 次
- StarNUMA: Mitigating NUMA Challenges with Memory PoolingAlbert Cho, Alexandros DaglisMICRO 2024 · 被引用 11 次
相关 Paper
- CTXNL: A Software-Hardware Co-designed Solution for Efficient CXL-Based Transaction ProcessingZhao Wang, Yiqi Chen, Cong Li, Yijin Guan 等ASPLOS 2025 · 被引用 9 次
- Systematic CXL Memory Characterization and Performance Analysis at ScaleJinshu Liu, Hamid Hadian, Yuyue Wang, Daniel S. Berger 等ASPLOS 2025 · 被引用 47 次
- Exploiting Burstiness to Improve Multi-Core Interference AnalysisJad Khatib, Dumitru Potop-Butucaru, Philippe BaufretonRTSS 2025
- Exploring Performance and Cost Optimization with ASIC-Based CXL MemoryYupeng Tang, Ping Zhou, Wenhui Zhang, Henry Hu 等EuroSys 2024 · 被引用 40 次
- CARINA: An Efficient CXL-Oriented Embedding Serving System for Recommendation ModelsPeiqi Yin, Qihui Zhou, Xiao Yan, Chao Wang 等SIGMOD 2025 · 被引用 2 次
