Zhijing Ye

Ziyu Hu, Zhiqing Zhong, Weijian Zheng et al.

The exponential growth of large language models has outpaced the capabilities of traditional CPU and GPU architectures due to the slowdown of Moore's Law. Dataflow AI accelerators present a promising alternative; however, there remains a lack of in-depth performance analysis and standardized benchmarking methodologies for LLM training. We introduce DABench-LLM, the first benchmarking framework designed for evaluating LLM workloads on dataflow-based accelerators. By combining intra-chip performance profiling and inter-chip scalability analysis, DABench-LLM enables comprehensive evaluation across key metrics such as resource allocation, load balance, and resource efficiency. The framework helps researchers rapidly gain insights into underlying hardware and system behaviors, and provides guidance for performance optimizations. We validate DABench-LLM on three commodity dataflow accelerators, Cerebras WSE-2, SambaNova RDU, and Graphcore IPU. Our framework reveals performance bottlenecks and provides specific optimization strategies, demonstrating its generality and effectiveness across a diverse range of dataflow-based AI hardware platforms.

Jiamin Wang, Zhijing Ye, Xiaodong Yu

Collective communication is a major bottleneck for multi-node GPU workloads in scientific computing and distributed deep learning, especially when inter-node bandwidth is limited. Although NCCL provides optimized GPU-centric collectives, large messages can still dominate end-to-end performance. Existing compression-enabled collective libraries either rely on MPI-based stacks that cannot fully exploit NCCL, omit entropy coding, or tightly couple full compressors with communication primitives, limiting compression ratio, flexibility, and communication-computation overlap. This paper presents NCCLZ, a compression-enabled GPU collectives that decouples quantization and entropy coding and integrates them at different layers of the stack. NCCLZ places quantization at the interface, embeds entropy coding into NCCL primitives, uses a lightweight device-side selector to choose coding strategies, and overlaps compression with communication to reduce exposed overhead. Experiments on scientific datasets, training gradients, and synthetic workloads show up to 9.65x speedup over NCCL and up to 3.34x improvement over prior compression-assisted collective libraries.

Zhiqing Zhong, Zhijing Ye, Jiamin Wang et al.

Closed-loop tool-using agents are increasingly evaluated in executable web, code, and micro-task environments, but benchmark reports often conflate workloads, action-generating drivers, and the evidence admitted for systems-facing claims. We present an executable benchmarking suite that makes these objects explicit under a shared evidence-admission contract. The suite connects WebArena Verified, a SWE-Gym slice with SWE-bench-compatible verification, and MiniWoB++ through common workload adapters, task manifests, event schemas, replay/freeze policy, declared drivers, and reporting pipelines. In the canonical release, the gate separates paper-facing evidence from preflight, fixture, smoke, and diagnostic rows while preserving non-admitted artifacts for audit and onboarding. The admitted evidence records latency, invalid-action behavior, patch-generation cost, verifier metadata, replay bindings, and provenance under one auditable contract. The gate is decision-relevant rather than merely clerical: in a separate WebArena Verified controller study, clean-baseline and medium live-stressed evaluation select different fixed controller variants under the same workload and admission contract. The release is scoped as a benchmarking suite and admitted evidence, not a new agent policy, model leaderboard, backend comparison, or autonomous SWE-bench solver.

Zhiqing Zhong, Zhijing Ye, Jian Zhang et al.

Static-graph LLM decoders provide predictable launches, fixed tensor shapes, and low submission overhead, but online decoding exposes highly irregular KV-cache behavior: request lengths differ, EOS events arrive asynchronously, and logical histories fragment over time. Dynamic runtimes recover flexibility through paged KV management and step-level scheduling, while static-graph executors often over-reserve memory and suffer burst-time latency outliers. This paper studies whether much of this variability can be absorbed below a fixed decode interface. We present KV-RM, a runtime design that regularizes KV-cache movement beneath a static-graph LLM decoder. KV-RM decouples logical KV histories from physical storage, tracks active KV state through a block pager, and materializes each decode step through a single committed descriptor. A merge-staged transport path coalesces non-contiguous KV mappings into a small number of large transfer groups before a fixed-shape attention kernel consumes them. Optional bounded far-history summaries can be enabled under the same interface, but the core design does not depend on them. On a 2-GPU NVIDIA A100 node, KV-RM improves mixed-length decoding throughput and tail latency relative to a static-graph baseline, reduces reserved KV memory across workload families, and removes severe burst-time latency spikes under production-trace replay. These results suggest that KV-cache movement, rather than kernel shape, can be an effective boundary for recovering runtime flexibility in static-graph LLM serving.