Abstract

The improvement of access latency to disks has been significantly
lagged behind the rapid technical advancement of other computing
resources, such as CPUs, memory bandwidths, and networks. The
performance bottleneck of "memory wall" has been shifted to "disk wall"
that is a serious bottleneck for many data-intensive applications.
One barrier to speedup data accesses in disks is the limited ability of
operating systems to exploit "sequential locality" that has a high
performance potential --- for the same amount of data, sequential
accesses are several orders of magnitude faster than random accesses in
disks.

We have designed and built a basic system infrastructure called
DiskSeen, which puts the disk layout information on the OS map. With
DiskSeen, we are able to exploit dual localities (DULO): temporal
locality based on workload execution patterns, and sequential locality
in disks.  Specifically, we present two new buffer management
techniques:  DULO-Caching and DULO-Prefetching. DULO-Caching can
effectively holds frequently used random accessed data in buffer cache
to avoid slow disk accesses, but timely replaces sequentially but not
very frequently accessed data to take advantage of fast sequential disk
accesses.  DULO-Prefetching adaptively preloads sequentially stored
data blocks in disks that belong to multiple files to buffer caches,
and significantly improve the prefetching efficiency.  We have
implemented DiskSeen with DULO-Caching and DULO-Prefetching in Linux
Kernel version 2.6.11, and evaluated their performance by various
data-intensive workloads. We show their effectiveness and low overheads
in a practical system environment.

================================
Bio of the Speaker

Xiaodong Zhang is the Robert M. Critchfield Professor in Engineering,
and Chairman of the Department of Computer Science and Engineering at
the Ohio State University.

His research interests cover a wide spectrum in the areas of high
performance and distributed systems. Several technical innovations and
research results from his team have been adopted or being developed in
commercial products and open source systems with direct impact to some
core computing operations, including the permutation memory
interleaving technique first in the Sun MicroSystems' UltraSPARC IIIi
processor and then in the Sun's dual-core Gemini Processor, the token
thrashing protection mechanism and the Clock-Pro page replacement
algorithm for memory management in the Linux Kernel and NetBSD.

Xiaodong Zhang was the Director of Advanced Computational Research
Program at the National Science Foundation, 2001-2004. He is the
associate Editor-in-Chief of IEEE Transactions on Parallel and
Distributed Systems, and is also serving on the Editorial Boards of
IEEE Transactions on Computers, IEEE Micro, and Journal of Parallel and
Distributed Computing.

He received his Ph.D. in Computer Science from University of Colorado
at Boulder.