|
|
|
Song Jiang
|
|
Research
My
research interests include operating system, file and storage system, and
high-performance computing.
GRA Opening
There
is one GRA (Graduate Research Assistantship) position available in the group
starting from the winter or fall of 2013 for new Ph.D students. Interested
students are welcome to contact Dr. Jiang with their CVs and transcripts.
Teaching
Representative and Recent Publications (see also Full publication list):
v
Xuechen Zhang, Ke Liu, Kei Davis, and Song Jiang, "iBridge: Improving Unaligned Parallel
File Access with Solid-State Drives", in Proceedings of
the IEEE International Parallel and Distributed Processing Symposium (IPDPS'13), Boston, MA, May, 2013.
v
Ke Liu, Xuechen Zhang, Kei Davis, and Song Jiang, "Synergistic Coupling of SSD and Hard
Disk for QoS-aware Virtual Memory", in Proceedings of
2013 IEEE International Symposium on Performance Analysis of Systems and
Software (ISPASS'13),
Austin, TX, April, 2013. [Best
Paper Candidate]
v
Berk Atikoglu, Yuehai Xu, Eitan Frachtenberg, Song Jiang, and Mike Paleczny, "Workload
Analysis of a Large-Scale Key-Value Store", in
Proceedings of the ACM SIGMETRICS/Performance 2012 Conference (SIGMETRICS’12), London, United Kingdom, June,
2012.
v
Xuechen Zhang, Kei Davis, and Song Jiang, "iTransformer: Using SSD to Improve
Disk Scheduling for High-performance I/O", in
Proceedings of the IEEE International Parallel and Distributed Processing
Symposium (IPDPS'12),
Shanghai, China, May, 2012.
v
Xuechen Zhang, Kei Davis, and Song Jiang, "Opportunistic Data-driven Execution of
Parallel Programs for Efficient I/O Services", in
Proceedings of the IEEE International Parallel and Distributed Processing
Symposium (IPDPS'12),
Shanghai, China, May, 2012.
v
Yizhe Wang, Kei Davis, Yuehai Xu and Song Jiang, "iHarmonizer: Improving the Disk
Efficiency of I/O-intensive Multithreaded Codes", in
Proceedings of the IEEE International Parallel and Distributed Processing
Symposium (IPDPS'12), Shanghai,
China, May, 2012.
v
Xuechen Zhang, Kei Davis, and Song Jiang, "QoS
Support for End Users of I/O-intensive Applications using Shared Storage
Systems", in Proceedings of the ACM/IEEE International
Conference for High Performance Computing, Networking, Storage, and Analysis (SC'11), Seattle, WA, November, 2011.
v
Xuechen Zhang, Yuehai Xu, and
Song Jiang, "YouChoose: A Performance Interface
Enabling Convenient and Efficient QoS Support for Consolidated Storage Systems",
in Proceedings of 27th IEEE Symposium on Massive Storage Systems and
Technologies (MSST'11),
Denver, CO, May, 2011. [Best
Student Paper Award]
v
Yuehai Xu and Song
Jiang, "A Scheduling Framework
that Makes any Disk Schedulers Non-work-conserving solely based on Request
Characteristics", in Proceedings of the 9th USENIX Conference on
File and Storage Technologies (FAST'11), San Jose, CA, February, 2011.
v
Xuechen Zhang, Kei Davis, and Song Jiang, "IOrchestrator: Improving the
Performance of Multi-node I/O Systems via Inter-Server Coordination",
in Proceedings of the ACM/IEEE International Conference for High Performance
Computing, Networking, Storage, and Analysis (SC'10), New
Orleans
v
Xuechen Zhang and Song Jiang, "InterferenceRemoval:
Removing Interference of Disk Access for MPI Programs through Data Replication",
in Proceedings of 24th ACM International Conference on Supercomputing (ICS'10), Tsukuba ,
Japan
v
Xiaoning Ding, Song
Jiang, and Xiaodong Zhang, "A Framework Making Any Replacement
Algorithms (Almost) Lock Contention Free", in
Proceedings of the 25th IEEE Int'l Conference on Data Engineering (ICDE’09), Shanghai, China, March, 2009.
v
Xiaoning Ding,
Song Jiang, Feng Chen, Kei Davis, and Xiaodong Zhang, "DiskSeen: Exploiting Disk Layout and
Access History to Enhance I/O Prefetch", in Proceedings of 2007
USENIX Annual Technical Conference (USENIX'07), Santa Clara
v
Song Jiang,
Xiaoning Ding, Feng Chen, Enhua Tan, and Xiaodong Zhang, "DULO: an Effective Buffer Cache
Management Scheme to Exploit both Temporal and Spatial Locality", in
Proceedings of the 4th USENIX Conference on File and Storage Technologies (FAST '05) , San Francisco , CA
v
Song Jiang, Feng
Chen, and Xiaodong Zhang, ``CLOCK-Pro:
An Effective Improvement of the CLOCK Replacement ", in Proceedings of
2005 USENIX Annual Technical Conference (USENIX'05), Anaheim CA ,
April, 2005.
v
Song Jiang and
Xiaodong Zhang, "LIRS: An Efficient
Low Inter-Reference Recency Set Replacement Policy to Improve Buffer Cache
Performance" in Proceedings of the ACM SIGMETRICS Conference on
Measurement and Modeling of Computer Systems (SIGMETRICS'02), Marina Del Rey, CA, June, 2002.
Book
Advanced
Operating Systems and Kernel Applications: Techniques and Technologies (pdf)
Impacts of Dr. Jiang’s Representative Work:
- LIRS is part of MySQL, the world’s most popular open source database, which are used by Google, Yahoo!, YouTube, and Wikipedia. (see the LIRS-related code in MySQL)
- The LIRS replacement algorithm has been officially incorporated in the MySQL’s most recent release (version 5.1), announced in November, 2008.
- Read the Documentation on the Implementation.
- This Sigmetrics paper and its journal paper describe the algorithm.
- Towards an O(1) VM, by Rik van Riel. (A Linux architect's view of LIRS in virtual memory)
- The LIRS caching algorithm has been adopted in Infinispan data grid platform.
- A JBoss Community article about LIRS in Infinispan.
- Clock-pro has been officially adopted in NetBSD. (See Clock-pro related code in NetBSD)
- ``CLOCK-Pro: an effective improvement
of the CLOCK replacement", Proceedings of 2005 USENIX Annual
Technical Conference (USENIX'05),
- A report about
its Linux kernel implementation.
- The Linux Kernel
Development Page of Clock-Pro.
- Clock-Pro patches
in 2.6.12 developed by Rik van Riel from the Red Hat, Inc..
- An
alternative Linux kernel implementation of Clock-Pro by Peter Zijlstra
- A Clock-pro patch is available for OpenLDAP (an open source implementation of the Lightweight Directory Access Protocol), implemented by Howard Chu, Symas Corp. (source code).
- Advanced Page Replacement Projects established by the Linux Memory Management Development Team.
- Clock-pro is referred as an example in the section of Linux and Academia in Book Professional Linux Kernel Architecture by Wolfgan Mauerer.
- The swap token algorithm has been officially adopted in Linux Kernel version 2.6. (See swap-token related code in Linux)
- ``Token-ordered
LRU: an effective page replacement policy and its implementation in Linux
systems", Performance Evaluation, Vol. 60, Issue 1-4,
2005, pp. 5-29
- Discussion on
the implementation of the algorithm on lwn.net
- The token algorithm is
described in the widely used Linux kernel
book Understanding
the Linux Kernel (3rd edition) by D. P. Bovet and M. Casati
- A detailed description of the token algorithm in Linux is presented in a section in Book Professional Linux Kernel Architecture by Wolfgan Mauerer.
- The token algorithm is referred as an example in the section of Linux and Academia in Book Professional Linux Kernel Architecture by Wolfgan Mauerer.
- The concept of ``false LRU pages" causing system thrashing is quoted in the Linux Memory Management Wiki.
Students:
·
Yuehai Xu (Ph.D student, Summer
internships at Google 2013, Microsoft Research 2013, Facebook 2011 and 2012)
·
Jianqian Ou (Ph.D student)
·
Somali Maasai Bayi (Undergraduate
student)
Alumni:
·
Ke Liu (Graduated in 2013 with ME degree.
Summer internship
at Los Alamos National Laboratory, 2012)
·
Xuechen Zhang
(Graduated in 2012 with Ph.d degree. Now at GaTech as a Postdoc researcher.
Summer internships at VMware and Los
Alamos National Laboratory, 2011)
·
Yizhe Wang (graduated in 2011 with ME degree.Now at VMware)
Research Supports:
v
Twelve servers
(Intel 8-core Xeon and 64GB DRAM) donated by Facebook.
v
NSF,
“Adaptively Applying Data-driven Execution Mode to Remove I/O Bottleneck
for Data-intensive Computing”, 2012-2014 (PI).
v
NSF,
“Enabling Dual-mode Execution for Removing I/O Bottleneck: A Highly
Applicable Design and Implementation”, 2011-2012 (PI).
v
NSF,
“CAREER: Building Virtual Devices with QoS Assurance in a Consolidated
Storage Infrastructure”, 2009-2014 (PI).
v
NSF,
“Collaborative Research: Algorithms Design and Systems Implementation to
Improve Buffer Management for I/O Data Accesses”, 2007-2010, (PI).
v
NSF, “CRI:
Reconfigurable High Performance Cluster Computing and Medical Engineering
Applications” 2007-2009 (co-PI).