Data and Instruction Uniformity in Minimal Multi-Threading

Teo Milanez 1 Sylvain Collange 2 Fernando Magno Quintão Pereira 3 Wagner Meira 1 Renato Ferreira 1
2 ALF - Amdahl's Law is Forever
Inria Rennes – Bretagne Atlantique , IRISA-D3 - ARCHITECTURE
3 Laboratório de Linguagens de Programação
DCC - UFMG - Departamento de Ciência da Computação [Minas Gerais]
Abstract : Simultaneous Multi-Threading (SMT) is a hardware model in which different threads share the same instruction fetching unit. This model is a compromise between high parallelism and low hardware cost. Minimal Multi-Threading (MMT) is a technique recently proposed to share instructions and execution between threads in a SMT machine. In this paper we propose new ways to explore redundancies in the MMT execution model. First, we propose and evaluate a new thread reconvergence heuristics that handles function calls better than previous approaches. Second, we demonstrate the existence of substantial regularity in inter-thread memory access patterns. We validate our results on the four data-parallel applications present in the PARSEC benchmark suite. The new thread reconvergence heuristics is, on the average, 82% more efficient than MMT's original reconvergence method. Furthermore, about 69% to 87% of all the memory addresses are either the same for all the threads, or are affine expressions of the thread identifier. This observation motivates the design of newly proposed hardware that benefits from regularity in inter-thread memory accesses.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [18 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00755273
Contributor : Sylvain Collange <>
Submitted on : Tuesday, November 20, 2012 - 6:57:25 PM
Last modification on : Wednesday, April 17, 2019 - 7:22:02 PM
Document(s) archivé(s) le : Thursday, February 21, 2013 - 12:36:24 PM

File

Milanez_DataInstructionUniform...
Files produced by the author(s)

Identifiers

Citation

Teo Milanez, Sylvain Collange, Fernando Magno Quintão Pereira, Wagner Meira, Renato Ferreira. Data and Instruction Uniformity in Minimal Multi-Threading. 24th International Symposium on Computer Architecture and High Performance Computing, Oct 2012, New-York, NY, United States. pp.270-277, ⟨10.1109/SBAC-PAD.2012.21⟩. ⟨hal-00755273⟩

Share

Metrics

Record views

652

Files downloads

288