Supporting High Performance I/O with Effective Caching and Prefetching
Department of Electrical and Computer Engineering
Wayne State University
Friday, March 20
11:00 AM - 12:00 PM
3105 Engineering Building
Host: Li Xiao
A large spectrum of data-intensive applications, ranging from small system tools such as CVS and grep, to terascale simulation applications that process huge amounts of scientific data, demand efficient I/O support. In almost all computing platforms, the ubiquitous hard disk remains the most cost-effective medium for on-line storage. While the growth of hard-disk capacity nicely matches the rapidly increasing demand for storage, its electromechanical nature is such that performance improvements lag painfully far behind that of processor performance. We continue to observe that the disk bottleneck is worsening in modern computer systems.
In this talk I will present our research on improving disk I/O performance through a better utilization of disk buffer cache. I will describe an integrated caching and prefetching scheme, called DiskSeen, that not only makes access patterns of applications exploitable by the buffer cache, but also makes the data layout of the disk visible and exploitable by the buffer cache. By making disk layout visible to the buffer cache, Diskseen provides functionalities that existing systems do not have. Examples includes random disk accesses being treated differently than sequential accesses so that disk accesses become more sequential, and prefetching being carried out directly at the disk level using history access information so that metadata and inter-file prefetching is enabled. Using Linux kernel implementations I demonstrate that this technique can significantly improve the performance of a wide variety of applications.
Dr. Song Jiang is an assistant professor at the ECE department of Wayne State University. He received his Ph.D in computer science at the College of William and Mary in 2004. After that he had been a postdoctoral researcher at Los Alamos National Laboratory for two years. He is the recipient of the 2009 NSF Faculty Early Career Development Award. His current research projects include the provisioning of storage resources, middleware/compiler-level I/O arrangement, and management of buffer cache for high scalability in multi-processor/core systems. Several of his proposed algorithms, such as Swap-token and Clock-pro, have been officially incorporated into the Linux and NetBSD kernels.