High Volume Computing: Identifying and Characterizing Throughput Oriented Workloads in Data Centers

摘要

For the first time, this paper systematically identifies three categories ofthroughput oriented workloads in data centers: services, data processingapplications, and interactive real-time applications, whose targets are toincrease the volume of throughput in terms of processed requests or data, orsupported maximum number of simultaneous subscribers, respectively, and we coina new term high volume computing (in short HVC) to describe those workloads anddata center computer systems designed for them. We characterize and compare HVCwith other computing paradigms, e.g., high throughput computing,warehouse-scale computing, and cloud computing, in terms of levels, workloads,metrics, coupling degree, data scales, and number of jobs or service instances.We also preliminarily report our ongoing work on the metrics and benchmarks forHVC systems, which is the foundation of designing innovative data centercomputer systems for HVC workloads.