Achieving supercomputer performane for neural net simulation with an array of digital signal processors
Urs Müller, B. Bäumle, P. Kohler et al. · IEEE Micro · 1992
Music, a digital signal processor (DSP)-based system with a parallel distributed-memory architecture that provides enormous computing power yet retains the flexibility of a general-purpose computer, is discussed. It is shown that Music reaches a peak performance of 2.7 Gflops at a significantly lower cost, power consumption, and space requirement than conventional supercomputers. The Music system hardware, programming, and backpropagation implementation are described.>
Cited by 31 works. doi:10.1109/40.166714
Authors
References
- Learning Internal Representations by Error Propagation
- Validity of the single processor approach to achieving large scale computing capabilities
- The Warp Computer: Architecture, Implementation, and Performance
- Neural network simulation at Warp speed: how we got 17 million connections per second
- The GF11 supercomputer
- The RAP: a ring array processor for layered network calculations
- An implementation of backpropagation learning on GF11, a large SIMD parallel computer
- Parallel simulation of multilayered neural networks on distributed-memory multiprocessors
- The Language of Fractals