Chapter

Languages and Compilers for Parallel Computing

Volume 7760 of the series Lecture Notes in Computer Science pp 127-142

UCIFF: Unified Cluster Assignment Instruction Scheduling and Fast Frequency Selection for Heterogeneous Clustered VLIW Cores

  • Vasileios PorpodasAffiliated withLancaster UniversitySchool of Informatics, University of Edinburgh
  • , Marcelo CintraAffiliated withLancaster UniversitySchool of Informatics, University of Edinburgh

* Final gross prices may vary according to local VAT.

Get Access

Abstract

Clustered VLIW processors are scalable wide-issue statically scheduled processors. Their design is based on physically partitioning the otherwise shared hardware resources, a design which leads to both high performance and low energy consumption. In traditional clustered VLIW processors, all clusters operate at the same frequency. Heterogeneous clustered VLIW processors however, support dynamic voltage and frequency scaling (DVFS) independently per cluster. Effectively controlling DVFS, to selectively decrease the frequency of clusters with a lot of slack in their schedule, can lead to significant energy savings.

In this paper we propose UCIFF, a new scheduling algorithm for heterogeneous clustered VLIW processors with software DVFS control, that performs cluster assignment, instruction scheduling and fast frequency selection simultaneously, all in a single compiler pass. The proposed algorithm solves the phase ordering problem between frequency selection and scheduling, present in existing algorithms. We compared the quality of the generated code, using both performance and energy-related metrics, against that of the current state-of-the-art and an optimal scheduler. The results show that UCIFF produces better code than the state-of-the-art, very close to the optimal across the mediabench2 benchmarks, while keeping the algorithmic complexity low.

Keywords

clustered VLIW heterogeneous DVFS scheduling phase-ordering