Skip to content
Book Open access

XTRA: Unifying Cache Coherence and Concurrency Control for Distributed Transactions in a CXL Pod

Sep 2026 · Proceedings of the ACM SIGOPS 32nd Symposium on Operating Systems Principles · pp. 269-285 · 0 citations · 75 references

TL;DR

XTRA is presented, a novel synchronization architecture that efficiently scales CXL distributed TRAnsactions by decoupling the synchronization state from individual data records, consolidating it into a compact directory within the HCC region, while keeping data in the large non-coherent pool.

Abstract

Distributed transactions on CXL memory pools allow hosts to directly access shared data, bypassing message-based coordination. However, the limited cross-host hardware cache coherence (HCC) in CXL hinders synchronization for consistency and concurrency control across large memory pools. To overcome this limitation, we present XTRA, a novel synchronization architecture that efficiently scales CXL distributed TRAnsactions. XTRA decouples the synchronization state from individual data records, consolidating it into a compact directory within the HCC region, while keeping data in the large non-coherent pool. Distributed transactions leverage this decoupled directory to coordinate concurrent access across the entire shared pool. To ensure efficient and consistent access, XTRA unifies cache coherence with concurrency control by exploiting their semantic redundancy, directly repurposing transactional conflict detection to guarantee cache freshness lazily upon read. Furthermore, by using software mutual exclusion to manage the directory, XTRA's decoupled synchronization generalizes to commodity CXL memory without HCC, while preserving upper-layer protocols. Extensive evaluations demonstrate that XTRA improves throughput by up to 15.6× over state-of-the-art systems and reduces P50/P99 latency by up to 4.0×/283.9×.

Read PDF

Similar papers

Book Open access Sep 2026

MiTDM: Eliminating False Conflicts in Scalable MVCC in Disaggregated Memory

MiTDM introduces a hierarchical block version chain that combines the benefits of array and chained structures, supporting dynamic version expansion while maintaining low-latency access, and achieves up to 80.1% higher throughput compared to FaRMV2 and 42.7% higher than Motor.

Ao-Xin Wei, Jin-Tian Wu, Jian Zhou et al. · 0 citations
Conference Sep 2026

Efficient Distributed Shared Memory System with Recall-Based Hybrid and Dynamic Coherence

Distributed Shared Memory (DSM) transparently provides a unified global address space across a cluster, simplifying the development and deployment of distributed applications. However, existing DSMs synchronize data at a static page, fixedblock, or object granularity, which cannot match the diverse and evolving spatial...

Rui-Zhe Huang, Ke Liang, Zi-Yuan He et al. · 0 citations
Book Open access Sep 2026

Borges: A Low-Latency Distributed Shared Log on a CXL Memory/SSD Hybrid

Borges is presented, to the authors' knowledge, the first fault-tolerant shared log whose data and ordering state reside in a DRAM/SSD hybrid device accessed with Compute Express Link (CXL).

Hao-Wei Chen, Yi-Ming Xiang, Zhi-Peng Jia et al. · 0 citations
Book Open access Sep 2026

Mitigating False Collisions in Partition-Based Concurrency Control

Modern storage systems must scale request processing across many cores, making efficient concurrency control essential. Data partitioning routes operations to execution contexts that own portions of system state, reducing synchronization overhead but potentially introducing false collisions: unnecessary serialization i...

Matthew Curtis-Maury, Sushrut Bhowmik, Grey Files et al. · 0 citations
Open access Sep 2026

A Study on the Impact of Data Replication Strategies on Performance in Distributed Database System

Data replication is a fundamental mechanism in distributed database systems, supporting fault tolerance, availability, and reliable data access across networked nodes. Cockroach DB is a cloud native, open-source distributed SQL database that relies on replication and consensus protocols to maintain consistency and re...

Famvie Egboson · 0 citations
#machine learning Preprint Sep 2026

Composable CXL Memory as a Kubernetes-Native Shared Memory for LLM Serving

A DRA driver is presented that makes composable CXL memory a schedulable cluster resource, and the resulting shared-memory tier for cross-node KV-cache reuse in LLM serving is evaluated, demonstrating memory disaggregation rather than prefill/decode disaggregation.

Hong-Jiang Fan, Ke-Rao Zhang, David Habinsky et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.