Refereed journal article or data article (A1)

A moving threads processor architecture MTPA

List of Authors: Forsell M, Leppänen V

Publisher: SPRINGER

Publication year: 2011

Journal: Journal of Supercomputing

Journal name in source: JOURNAL OF SUPERCOMPUTING

Journal acronym: J SUPERCOMPUT

Number in series: 1

Volume number: 57

Issue number: 1

Number of pages: 15

ISSN: 0920-8542


Moving threads is a new kind of approach for multicore processor architectures. Traditionally, each thread stays in the core where it is created, and data is moved from the main memory via caches to each core and thread. In the moving threads approach, each core can access only a certain portion of the main memory via its local memory block, and thus extremely lightweight threads are moved between the cores. As a consequence, all kinds of cache coherence problems and need for read reply messages are eliminated. Also Lamport's sequential consistency of shared memory multiprocessor systems is achieved for free. In this paper, we propose a processor architecture (MTPA) for the moving threads paradigm. We describe the overall structure, operation, instruction set, and thread management mechanism as well as evaluate the proposed architecture with different functional unit settings with simulations and give early silicon area and power consumption estimates.

Last updated on 2021-24-06 at 10:25