Paper Info
Title
Improving the Operation Autonomy of SIMD Processing Elements by Using Guarded Instructions and Pseudo Branches
Abstract
This paper presents a novel method for improving the operation autonomy of the processing elements (PE) of SIMD-like machines. By combining guarded instructions and pseudo branches it is possible to achieve higher operation autonomy and higher instruction level parallelism than in previous SIMD/ASIMD architectures. The paper shows that it is feasible to avoid most branches and it is also possible to emulate conditional execution on the processing elements, either by using guarded instructions or by using pseudo branches, thus avoiding unnecessary intervention by the array control unit in data-dependant computations. Pseudo branches are used when it is not possible to use guarded instructions. Additionally, they alsosupport the implementation of complex nested if-then-else constructs, improving the execution of irregular data-parallel applications. The paper also shows that the simplicity of the method allows it to be implemented both in fine-grain and coarse-grain SIMD/ASIMD architectures because it does not require significant additional silicon area. Finally, it is shown that pseudo branches can be used to control the power saving of those processing elements that have instructions nullified.
Year
DOI
Venue
2002
10.1109/DSD.2002.1115363
DSD
Keywords
Field
DocType
pseudo branch,asimd architecture,higher instruction level parallelism,array control unit,higher operation autonomy,simd processing elements,conditional execution,novel method,guarded instruction,pseudo branches,coarse-grain simd,guarded instructions,operation autonomy,processing element,silicon,reduced instruction set computing,instruction level parallelism,computer architecture,parallel processing,application software,acceleration,digital signal processing
Instruction-level parallelism,Digital signal processing,Computer science,Parallel computing,SIMD,Reduced instruction set computing,Control unit,Acceleration,Application software,Computation
Conference
ISBN
Citations 
PageRank 
0-7695-1790-0
14
1.00
References 
Authors
10
3
Name
Order
Citations
PageRank
M. L. Anido1282.80
Alexander Paar2446.69
Nader Bagherzadeh31674182.54