Measuring the performance of multimedia instruction sets

Slingerland, Nathan T.; Smith, Alan Jay

doi:10.1109/tc.2002.1047756

Cited by 59 publications

(42 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Slingerland and Smith [13] study the performance of various SIMD instruction sets on several multimedia kernels. Interestingly (and independently), they also found two factors that limit the performance of current media ISA extensions: storage formats which are insufficient for computation (necessitating conversion overhead), and nonunit strides.…”

Section: Related Workmentioning

confidence: 99%

The CSI multimedia architecture

Cheresiz

Juurlink

Vassiliadis

et al. 2005

IEEE Trans. VLSI Syst.

View full text Add to dashboard Cite

Abstract-An instruction set extension designed to accelerate multimedia applications is presented and evaluated. In the proposed complex streamed instruction (CSI) set, a single instruction can process vector data streams of arbitrary length and stride and combines complex memory accesses (with implicit prefetching), program control for vector sectioning, and complex computations on multiple data in a single operation. In this way, CSI eliminates overhead instructions (such as instructions for data sectioning, alignment, reorganization, and packing/unpacking) often needed in applications utilizing MMX-like extensions and accelerates key multimedia kernels. Simulation results demonstrate that a superscalar processor extended with CSI outperforms the same processor enhanced with Sun's VIS extension by a factor of up to 7.77 on key multimedia kernels and by up to 35% on full applications.

show abstract

Section: Related Workmentioning

confidence: 99%

The CSI multimedia architecture

Cheresiz

Juurlink

Vassiliadis

et al. 2005

IEEE Trans. VLSI Syst.

View full text Add to dashboard Cite

show abstract

“…All experiments were conducted on a 2.8G Pentium 4E Processor and 1G memory system with Redhat 9.0. The benchmark we use is Accelerating Suite of Berkeley Multimedia Workload (ASBMW) [4] [5]. We also compare the results of our method with ICC [10], the Intel compiler that has the state of the art vectorization techniques.…”

Section: Resultsmentioning

confidence: 99%

“…The corresponding operations are heavily used in real-life multimedia applications. And after manual vectorization, they contribute to almost all speedups [4] [5]. This fact shows that the recognition of these ISA-supported multimedia typical operations should be the focus of compiler support for MMEs.…”

Section: Key Points In Fully Utilizing Mmesmentioning

confidence: 94%

“…Experimental results show that we can vectorize many operations in real-life multimedia applications and the performances are quite satisfactory. We achieved a 10.86% average speedup for Accelerating Suite of Berkeley Multimedia Workload [4] [5]. Compared with the vectorization ability of Intel C Compiler (ICC) version 8 [10], our compiler outperforms it by about 8% on average.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Boosting the Performance of Multimedia Applications Using SIMD Instructions

Jiang

Mei

Huang³

et al. 2005

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Modern processors' multimedia extensions (MME) provide SIMD ISAs to boost the performance of typical operations in multimedia applications. However, automatic vectorization support for them is not very mature. The key difficulty is how to vectorize those SIMD-ISA-supported idioms in source code in an efficient and general way. In this paper, we introduce a powerful and extendable recognition engine to solve this problem, which only needs a small amount of rules to recognize many such idioms and generate efficient SIMD instructions. We integrated this engine into the classic vectorization framework and obtained very good performance speedup for some real-life applications.

show abstract

“…However, most SIMD media extensions have a limited memory architecture which provides access to only contiguous data in memory, with strong alignment restrictions and a weak support for partial load and stores [3], [4]. These architectures, either do not provide any hardware support for unaligned accesses or provide it but at the expense of a big performance penalty.…”

Section: Introductionmentioning

confidence: 99%

Performance Impact of Unaligned Memory Operations in SIMD Extensions for Video Codec Applications

Álvarez¹,

Salamí²,

Ramírez

et al. 2007

2007 IEEE International Symposium on Performance Analysis of Systems &Amp; Software

View full text Add to dashboard Cite

Abstract-Although SIMD extensions are a cost effective way to exploit the data level parallelism present in most media applications, we will show that they had have a very limited memory architecture with a weak support for unaligned memory accesses. In video codec, and other applications, the overhead for accessing unaligned positions without an efficient architecture support has a big performance penalty and in some cases makes vectorization counter-productive. In this paper we analyze the performance impact of extending the Altivec SIMD ISA with unaligned memory operations. Results show that for several kernels in the H.264/AVC media codec, unaligned access support provides a speedup up to 3.8X compared to the plain SIMD version, translating into an average of 1.2X in the entire application. In addition to providing a significant performance advantage, the use of unaligned memory instructions makes programming SIMD code much easier both for the manual developer and the autovectorizing compiler.

show abstract

Measuring the performance of multimedia instruction sets

Cited by 59 publications

References 21 publications

The CSI multimedia architecture

The CSI multimedia architecture

Boosting the Performance of Multimedia Applications Using SIMD Instructions

Performance Impact of Unaligned Memory Operations in SIMD Extensions for Video Codec Applications

Contact Info

Product

Resources

About