English

Transaction Level Analysis for a Clustered and Hardware-Enhanced Task Manager on Homogeneous Many-Core Systems

Distributed, Parallel, and Cluster Computing 2015-02-11 v1

Abstract

The increasing parallelism of many-core systems demands for efficient strategies for the run-time system management. Due to the large number of cores the management overhead has a rising impact to the overall system performance. This work analyzes a clustered infrastructure of dedicated hardware nodes to manage a homogeneous many-core system. The hardware nodes implement a message passing protocol and perform the task mapping and synchronization at run-time. To make meaningful mapping decisions, the global management nodes employ a workload status communication mechanism. This paper discusses the design-space of the dedicated infrastructure by means of task mapping use-cases and a parallel benchmark including application-interference. We evaluate the architecture in terms of application speedup and analyze the mechanism for the status communication. A comparison versus centralized and fully-distributed configurations demonstrates the reduction of the computation and communication management overhead for our approach.

Keywords

Cite

@article{arxiv.1502.02852,
  title  = {Transaction Level Analysis for a Clustered and Hardware-Enhanced Task Manager on Homogeneous Many-Core Systems},
  author = {Daniel Gregorek and Robert Schmidt and Alberto Garcia-Ortiz},
  journal= {arXiv preprint arXiv:1502.02852},
  year   = {2015}
}

Comments

Presented at HIP3ES, 2015 (arXiv: 1501.03064)

R2 v1 2026-06-22T08:26:26.216Z