A bandwidth-efficient architecture for media processing
Scott Rixner, W.J. Dally, Ujval J. Kapasi et al. · 1998
Media applications are characterized by large amounts of available parallelism, little data reuse, and a high computation to memory access ratio. While these characteristics are poorly matched to conventional microprocessor architectures, they are a good fit for modern VLSI technology with its high arithmetic capacity but limited global bandwidth. The stream programming model, in which an application is coded as streams of data records passing through computation kernels, exposes both parallelism and locality in media applications that can be exploited by VLSI architectures. The Imagine architecture supports the stream programming model by providing a bandwidth hierarchy tailored to the demands of media applications. Compared to a conventional scalar processor. Imagine reduces the global register and memory bandwidth required by typical applications by factors of 13 and 21 respectively. This bandwidth efficiency enables a single chip Imagine processor to achieve a peak performance of 16.2GFLOPS (single-precision floating point) and sustained performance of up to 8.5GFLOPS on media processing kernels.
Cited by 127 works. doi:10.1109/micro.1998.742118
Authors
- Scott Rixner
- W.J. Dally
- Ujval J. Kapasi
- Brucek Khailany
- Abelardo Lopez-Lagunas
- Peter R. Mattson
- John D. Owens
References
- MMX technology extension to the Intel architecture
- Computer Graphics — Principles and Practice
- Subword parallelism with MAX-2
- VIS speeds new media processing
- How multimedia workloads will change processor design
- Point Sample Rendering
- Cheops: a reconfigurable data-flow system for video processing
- Challenges to combining general-purpose and multimedia processors
- Available parallelism in video applications