
Efficient Processor Allocation Strategies for Mesh Multicomputers
Performance Evaluation of Processor Allocation Strategies for Mesh Interconnection Networks
Versandkostenfrei!
Versandfertig in 6-10 Tagen
51,99 €
inkl. MwSt.
PAYBACK Punkte
26 °P sammeln!
We have identified the limitations of the existing work in the area of processor allocation in mesh-connected multicomputers. These limitations form the motives behind undertaking such a research. We can summarise the main contributions of this research into three main points. The first part of this research presents an efficient contiguous processor allocation algorithm, referred to as Turning Busy List (or TBL for short), for 3D mesh-connected multicomputers that overcome the limitations of the existing strategies suggested previously for the 3D mesh networks. The second part of this researc...
We have identified the limitations of the existing
work in the area of processor allocation in mesh-
connected multicomputers. These limitations form the
motives behind undertaking such a research. We can
summarise the main contributions of this research
into three main points. The first part of this
research presents an efficient contiguous processor
allocation algorithm, referred to as Turning Busy
List (or TBL for short), for 3D mesh-connected
multicomputers that overcome the limitations of the
existing strategies suggested previously for the 3D
mesh networks. The second part of this research
proposes a new non-contiguous processor allocation
algorithm, referred to as Greedy Available Busy List
(or GABL for short), for the 2D mesh-connected
multicomputer. And finally, we study the impact of
heavy-tailed job execution times on the performance
of the existing contiguous allocation strategies for
3D mesh-connected multicomputers, including the ones
proposed in this research.
work in the area of processor allocation in mesh-
connected multicomputers. These limitations form the
motives behind undertaking such a research. We can
summarise the main contributions of this research
into three main points. The first part of this
research presents an efficient contiguous processor
allocation algorithm, referred to as Turning Busy
List (or TBL for short), for 3D mesh-connected
multicomputers that overcome the limitations of the
existing strategies suggested previously for the 3D
mesh networks. The second part of this research
proposes a new non-contiguous processor allocation
algorithm, referred to as Greedy Available Busy List
(or GABL for short), for the 2D mesh-connected
multicomputer. And finally, we study the impact of
heavy-tailed job execution times on the performance
of the existing contiguous allocation strategies for
3D mesh-connected multicomputers, including the ones
proposed in this research.