Abstract—The layout of traditional disks is optimized for file systems with large sequential access. For these sequential workloads, cost of seeks and rotational latencies are amortized on the transfer of a large amount of data to/ from the disk. Each access from a file system workloads, however incurs a time consuming seek and rotational latency that dominate the request response time, resulting in very less data rate. In the designed model to enhance the speed of data accessing from the hard disk and there by improving the over all performance of the system, we distribute one file in to the different platters available in the hard disk. For this, first split the file in to the number of parts based on the block size and the number of platters available in the hard disk by the file splitting storage algorithm. The first part is stored on the first storage surface. Then, the second part is stored on the same cylindrical location of the second storage surface. This process is repeated for the remaining parts of the file. During the reading process, place the head dedicated to the storage surface to the correct location. Then perform the reading simultaneously. The access time can be reduced drastically. The main intention behind the designs model is to supply the data as fast as possible to the RAM. This model is helpful for enhancing the virtual memory and paging mechanism by creating an illusion of the presence of data in the RAM, so that user can execute his program more efficiently with less amount of RAM.
Index Terms—Hard disk drive, internal data transfer, internal fragmentation, response time.
Mr. Shihabudheen P M was with the department of computer science and engineering, SCMS school of engineering and technology, Cochin, India. He is now with the Broadcast Business Unit, Tata Elxsi Limited, Trivandrum, India (e-mail: shihabudheen27@gmail.com).
Cite: Shihabudheen P M, "Data Distribution and Improving Disk Performance for Faster Memory Access," International Journal of Computer Theory and Engineering vol. 4, no. 2, pp. 181-184, 2012.
Copyright © 2008-2024. International Association of Computer Science and Information Technology. All rights reserved.
