Espresso: Constructing Cost-Efficient CXL JBOF via Inter-SSD Computing Resource Sharing
Shushu Yi, Yuda An, Li Peng, Xiurui Pan, Qiao Li, Jieming Yin, Guangyan Zhang, Wenfei Wu, Chenxi Wang, Diyu Zhou, Zhenlin Wang, Xiaolin Wang
Abstract
Enterprise SSDs integrate substantial computing resources (e.g., ARM processor and onboard DRAM) to handle I/O bursts. However, these resources significantly raise SSD monetary cost and suffer severely underutilized in JBOF deployments due to the sporadic nature of I/O bursts. In this paper, we present Espresso , a cost-efficient JBOF design that provisions only moderate computing resources per SSD at low monetary cost, while delivering demanded I/O performance through efficient inter-SSD resource sharing. Specifically, Espresso first disaggregates SSD architecture into functionally distinct components, enabling fine-grained SSD internal resource management. Espresso then employs a decentralized scheme to manage these disaggregated resources and harvests the computing resources of idle SSDs to assist busy SSDs in handling I/O bursts. This idea is facilitated by the cache-coherent CXL fabric, with which the busy SSDs can directly utilize the harvested computing resources to accelerate metadata processing. The evaluation results show that Espresso can improve SSD resource utilization by 50.4% and reduce monetary cost by 19.0% with negligible performance degradation, compared to the state-of-the-art JBOF designs.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Your agent calls
Luneget_paper_fulltext
Free to start. No credit card required.
Terminal
Install the CLIlune papers fulltext d462a3c7-053e-4cac-965d-1a72c64f8aedBuilds on42
- Pond: CXL-Based Memory Pooling Systems for Cloud PlatformsHuaicheng Li, Daniel S. Berger, Lisa Hsu, Daniel Ernst et al.ASPLOS 2023 · 328 citations
- ZNS: Avoiding the Block Interface Tax for Flash-based SSDsMatias Bjørling, Abutalib Aghayev, Hans Holmberg, Aravind Ramesh et al.USENIX ATC 2021 · 221 citations
- Demystifying CXL Memory with Genuine CXL-Ready Systems and DevicesYan Sun, Yifan Yuan, Zeduo Yu, Reese Kuper et al.MICRO 2023 · 133 citations
- Providing SLOs for Resource-Harvesting VMs in Cloud PlatformsPradeep Ambati, Iñigo Goiri, Felipe Vieira Frujeri, Alper Gun et al.OSDI 2020 · 101 citations
- Overcoming the Memory Wall with CXL-Enabled SSDsShao-Peng Yang, Minjae Kim, Sanghyun Nam, Juhyung Park et al.USENIX ATC 2023 · 75 citations
Related papers
- XHarvest: Rethinking High-Performance and Cost-Efficient SSD Architecture with CXL-Driven HarvestingLi Peng, Wenbo Wu, Shushu Yi, Xianzhang Chen et al.ISCA 2025 · 2 citations
- SkyByte: Architecting an Efficient Memory-Semantic CXL-based SSD with OS and Hardware Co-designHaoyang Zhang, Yuqi Xue, Yirui Eric Zhou, Shaobo Li et al.HPCA 2025 · 8 citations
- Scalio: Scaling up DPU-based JBOF Key-value Store with NVMe-oF Target OffloadXun Sun, Mingxing Zhang, Yingdi Shan, Kang Chen et al.OSDI 2025 · 2 citations
- Stream-Based Data Placement for Near-Data Processing with Extended MemoryYiwei Li, Boyu Tian, Yi Ren, Mingyu GaoMICRO 2024 · 5 citations
- IOCost: block IO control for containers in datacentersTejun Heo, Dan Schatzberg, Andrew Newell, Song Liu et al.ASPLOS 2022 · 22 citations
