Affiliation:
1. Stanford University, Stanford, CA, USA
2. Hicamp Systems, Menlo Park, CA, USA
Abstract
Due to their high volume, general-purpose processors, and now chip multiprocessors (CMPs), are much more cost effective than ASICs, but lag significantly in terms of performance and energy efficiency. This paper explores the sources of these performance and energy overheads in general-purpose processing systems by quantifying the overheads of a 720p HD H.264 encoder running on a general-purpose CMP system. It then explores methods to eliminate these overheads by transforming the CPU into a specialized system for H.264 encoding. We evaluate the gains from customizations useful to broad classes of algorithms, such as SIMD units, as well as those specific to particular computation, such as customized storage and functional units.The ASIC is 500x more energy efficient than our original four-processor CMP. Broadly applicable optimizations improve performance by 10x and energy by 7x. However, the very low energy costs of actual core ops (100s fJ in 90nm) mean that over 90% of the energy used in these solutions is still "overhead". Achieving ASIC-like performance and efficiency requires algorithm-specific optimizations. For each sub-algorithm of H.264, we create a large, specialized functional unit that is capable of executing 100s of operations per instruction. This improves performance and energy by an additional 25x and the final customized CMP matches an ASIC solution's performance within 3x of its energy and within comparable area.
Publisher
Association for Computing Machinery (ACM)
Reference40 articles.
1. Horowitz M.; Alon E.; Patil D.; Naffziger S.; Kumar R.; Bernstein K. "Scaling Power and the Future of CMOS " 20th Int'l Conference on VLSI Design 2007 held jointly with 6th Int'l Conference on Embedded Systems p. 23 6--10 Jan. 2007 10.1109/VLSID.2007.140 Horowitz M.; Alon E.; Patil D.; Naffziger S.; Kumar R.; Bernstein K. "Scaling Power and the Future of CMOS " 20th Int'l Conference on VLSI Design 2007 held jointly with 6th Int'l Conference on Embedded Systems p. 23 6--10 Jan. 2007 10.1109/VLSID.2007.140
2. Chip multi-processor generator
3. Analysis and architecture design of an HDTV720p 30 frames/s H.264/AVC encoder
4. A design environment for high-throughput low-power dedicated signal processing systems
Cited by
189 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. ADS-CNN: Adaptive Dataflow Scheduling for lightweight CNN accelerator on FPGAs;Future Generation Computer Systems;2024-09
2. SLIDEX: A Novel Architecture for Sliding Window Processing;Proceedings of the 38th ACM International Conference on Supercomputing;2024-05-30
3. Enabling Efficient Hardware Acceleration of Hybrid Vision Transformer (ViT) Networks at the Edge;2024 IEEE International Symposium on Circuits and Systems (ISCAS);2024-05-19
4. A 16.41 TOPS/W CNN Accelerator with Event-Based Layer Fusion for Real-Time Inference;2024 IEEE International Symposium on Circuits and Systems (ISCAS);2024-05-19
5. Fused Functional Units for Area-Efficient CGRAs;2024 25th International Symposium on Quality Electronic Design (ISQED);2024-04-03