arXiv subjects

Xinjun Yang

Publications and source records attributed to Xinjun Yang.

4 recordsLinked to original sources

arXiv2026

SAC: Disaggregated KV Cache System for Sparse Attention LLMs with CXL

The scaling of LLMs toward long-context inference has shifted the primary serving system bottleneck from computation to memory capacity. Traditional solutions for dense attention models rely on RDMA-based disaggregated memory pools, which perform coarse-grained fetching of the entire prefix KV cache from remote storage to local memory before decoding. However, this approach is fundamentally inefficient for emerging sparse attention models. While only a small fraction of KV entries are active during decoding, these systems still fetch the full KV cache locally, leading to severe transmission bottlenecks and local memory wastage. To address this, we propose SAC, the first efficient disaggregated KV cache system optimized for sparse attention models. By leveraging the low-latency, cache-line granularity load/store semantics of Compute Express Link (CXL), SAC fetches only the required top-k KV entries on demand during inference. Evaluations on DeepSeek-V3.2 using SGLang show that SAC achieves 2.1x higher throughput, 9.7x lower TTFT, and 1.8x lower TBT compared to RDMA-based baselines, establishing CXL-based disaggregation as the superior infrastructure for emerging sparse attention models.

cs.DC↗

arXiv2025

PolarStore: High-Performance Data Compression for Large-Scale Cloud-Native Databases

In recent years, resource elasticity and cost optimization have become essential for RDBMSs. While cloud-native RDBMSs provide elastic computing resources via disaggregated computing and storage, storage costs remain a critical user concern. Consequently, data compression emerges as an effective strategy to reduce storage costs. However, existing compression approaches in RDBMSs present a stark trade-off: software-based approaches incur significant performance overheads, while hardware-based alternatives lack the flexibility required for diverse database workloads. In this paper, we present PolarStore, a compressed shared storage system for cloud-native RDBMSs. PolarStore employs a dual-layer compression mechanism that combines in-storage compression in PolarCSD hardware with lightweight compression in software. This design leverages the strengths of both approaches. PolarStore also incorporates database-oriented optimizations to maintain high performance on critical I/O paths. Drawing from large-scale deployment experiences, we also introduce hardware improvements for PolarCSD to ensure host-level stability and propose a compression-aware scheduling scheme to improve cluster-level space efficiency. PolarStore is currently deployed on thousands of storage servers within PolarDB, managing over 100 PB of data. It achieves a compression ratio of 3.55 and reduces storage costs by approximately 60%. Remarkably, these savings are achieved while maintaining performance comparable to uncompressed clusters.

cs.DC↗

arXiv2025

Beluga: A CXL-Based Memory Architecture for Scalable and Efficient LLM KVCache Management

The rapid increase in LLM model sizes and the growing demand for long-context inference have made memory a critical bottleneck in GPU-accelerated serving systems. Although high-bandwidth memory (HBM) on GPUs offers fast access, its limited capacity necessitates reliance on host memory (CPU DRAM) to support larger working sets such as the KVCache. However, the maximum DRAM capacity is constrained by the limited number of memory channels per CPU socket. To overcome this limitation, current systems often adopt RDMA-based disaggregated memory pools, which introduce significant challenges including high access latency, complex communication protocols, and synchronization overhead. Fortunately, the emerging CXL technology introduces new opportunities in KVCache design. In this paper, we propose Beluga, a novel memory architecture that enables GPUs and CPUs to access a shared, large-scale memory pool through CXL switches. By supporting native load/store access semantics over the CXL fabric, our design delivers near-local memory latency, while reducing programming complexity and minimizing synchronization overhead. We conduct a systematic characterization of a commercial CXL switch-based memory pool and propose a set of design guidelines. Based on Beluga, we design and implement Beluga-KVCache, a system tailored for managing the large-scale KVCache in LLM inference. Beluga-KVCache achieves an 89.6% reduction in Time-To-First-Token (TTFT) and 7.35x throughput improvement in the vLLM inference engine compared to RDMA-based solutions. To the best of our knowledge, Beluga is the first system that enables GPUs to directly access large-scale memory pools through CXL switches, marking a significant step toward low-latency, shared access to vast memory resources by GPUs.

cs.DC↗

arXiv2023

PolarDB-IMCI: A Cloud-Native HTAP Database System at Alibaba

Cloud-native databases have become the de-facto choice for mission-critical applications on the cloud due to the need for high availability, resource elasticity, and cost efficiency. Meanwhile, driven by the increasing connectivity between data generation and analysis, users prefer a single database to efficiently process both OLTP and OLAP workloads, which enhances data freshness and reduces the complexity of data synchronization and the overall business cost. In this paper, we summarize five crucial design goals for a cloud-native HTAP database based on our experience and customers' feedback, i.e., transparency, competitive OLAP performance, minimal perturbation on OLTP workloads, high data freshness, and excellent resource elasticity. As our solution to realize these goals, we present PolarDB-IMCI, a cloud-native HTAP database system designed and deployed at Alibaba Cloud. Our evaluation results show that PolarDB-IMCI is able to handle HTAP efficiently on both experimental and production workloads; notably, it speeds up analytical queries up to $\times149$ on TPC-H (100 $GB$). PolarDB-IMCI introduces low visibility delay and little performance perturbation on OLTP workloads (< 5%), and resource elasticity can be achieved by scaling out in tens of seconds.

cs.DB↗