Learning process is essential for good performance when a neural network is applied to a practical application. The backpropagation algorithm is a well-known learning method widely used in most neural networks. However, since the backpropagation algorithm is time-consuming, much research have been done to speed up the process. The block backpropagation algorithm, which seems to be more efficient than the backpropagation, is recently proposed by Coetzee in [2]. In this paper, we propose an efficient parallel algorithm for the block backpropagation method and its performance model in mesh-connected parallel computer systems. The proposed algorithm adopts master-slave model for weight broadcasting and data parallelism for computation of weights. In order to validate our performance model, a neural network is implemented for printed character recognition application in the TiME which is a prototype parallel machine consisting of 32 transputers connected in mesh topology. It is shown that speedup by our performance model is very close to that by experiments.

There is a growing demand for high reliability beyond what current RAID can provide and there are various levels of user demand for data reliability. An efficient data placement scheme called RM2 has been proposed in [10], which makes a disk array system resistant to double disk failures. In this paper, we consider how to choose an optimal striping unit for RM2 particularly when no workload information is available except read/write ratio. For experimental purposes, we develop a disk array simulator incorporating RM2 as one of the data placement schemes including other schemes of RAID levels. In the case of disk read operations, it is shown that RM2 has an optimal striping unit of 4/3T for large requests and 8/3T for small requests, where T represents the size of a single track. We have also shown that, if any disk write operations are involved, an optimal striping unit becomes 1/3T for large requests and 8/3T for small requests.

Disk arrays are widely accepted as a disk subsystem for video servers due to its high throughput as well as high concurrency. RAID-like disk arrays are usually managed in either RAID level 3 (a request is handled by all the disks in the system) or RAID level 5 (a request is handled by some number of disks subject to the request size) when they are used in video servers, i. e. , either only one video stream is handled at a time in RAID level 3 or a certain number of video streams are handled independently at the same time in RAID level 5. Note that RAID level 3 is inappropriate to handle large number of video streams and RAID level 5 is inefficient to handle multiple video streams since handling continuous video streams is inherently synchronous operation. In this paper, we propose a new video data layout scheme called region-based layout and synchronous management of RAID5 called synchronous RAID5 for disk array used in video servers. It is shown that we can reduce the amount of buffers required to support a given number of video requests by integrating our region-based layout with synchronous RAID5 scheme. Group Sweeping Scheduling (GSS) is used as a basic disk scheduling. We have shown through analysis that our proposed scheme is superior to the existing schemes in the respect of the buffer requirements.