Search papers, labs, and topics across Lattice.
This paper introduces CC-Bench, a benchmark suite designed to evaluate communication compression in distributed HPC and LLM workloads under realistic conditions. By utilizing declarative application-environment modeling and combining function-level interception with hardware counter monitoring, CC-Bench effectively assesses per-phase latency, hardware utilization, and numerical fidelity. The evaluation of three communication libraries reveals critical accuracy-performance trade-offs, providing insights for optimizing deployment in high-performance environments.
Communication compression can significantly enhance performance in HPC and LLM workloads, but existing benchmarks fail to capture its true potential鈥擟C-Bench changes that narrative.
Distributed HPC and LLM workloads increasingly require efficient communication for scalability, yet growing data movement has become a major performance bottleneck. Communication compression can reduce this overhead and complement execution-level optimizations, but its benefits remain difficult to assess because existing benchmarks lack support for diverse backends, realistic datasets, application-specific accuracy metrics, and overlap-induced resource contention. We present CC-Bench, a lightweight, extensible, and application-oriented benchmark suite for evaluating communication compression under realistic execution conditions. CC-Bench uses declarative application-environment modeling to decouple profiling logic from communication libraries, datasets, and fidelity metrics, enabling portable cross-library evaluation. It further combines function-level interception and hardware counter monitoring to characterize per-phase latency, hardware utilization, numerical fidelity, and computation interference. With representative datasets from HPC and LLM workloads, CC-Bench evaluates three compression-enabled communication libraries on CPU and GPU clusters, revealing accuracy-performance trade-offs and bottlenecks to guide practical deployment and optimization.
