Xubin (Ben) He, Ph.D
Department of Computer & Information Sciences
College of Science and Technology, Temple University
Electrical & Computer Engineering, University of Rhode Island, 2002
M.A: Computer Science, Huazhong U. of Sci. & Tech., 1997
Science, Huazhong U. of Sci. & Tech., 1995
STAR Lab News
-  Pandemic due to COVID-19, most classes move to online, so do the conferences (IPDPS, ICPP, USENIX ATC, etc). Best Wishes!
- [01-28-2020] Our two papers are accepted for presentation at IEEE IPDPS conference, New Orleans, May 18-22, 2020.
- [01-06-2020] Our research projects on big data and IoT are featured in CIS news letter.
- [08-26-2019] Welcome new PhD student Peiyu Liang to join our lab!
- [12-11-2018] Our paper "GearDB: A GC-free Key-Value Store on HM-SMR Drives with Gear Compaction" has been accepted for presentation at the USENIX Conf. on File and Storage Technologies (FAST'2019), acceptance rate: 26/145=18%. USENIX FAST is a premier conference in data storage systems.
- [06-22-2018] NSF CCF Award: 2018-2021, $460k. Our NSF proposal "SHF:Small: Collaborative Research: Understanding, Modeling, and System Support for HPC Data Reduction" was selected for funding.The goal of this project is to address a number of critical gaps in using data reduction for HPC-based science missions. This is a collaborative project with NJIT.
- [04-30-2018] Yuhua (co-supervised with Dr. Weijun Xiao@VCU) successfully defended his dissertation. He is going to join DataDirect Networks (DDN). Congratulations, Dr. Guo!
- [04-18-2018] Our paper "RAFI: Risk-Aware Failure Identification to Improve the RAS in Erasure-coded Data Centers" has been accepted for presentation at the 2018 USENIX Annual Technical Conference (ATC'2018), acceptance rate: 76/378=20%. This is a joint work with collaborators at HUST. USENIX ATC is a premier conference in computer systems.
- [[January 2018] Congratulations to Ting and Wenjie who both received USENIX travel grants to attend the FAST'18 conference.
1925 N. 12th Street, Philadelphia, PA 19122-1801
Office: Room 352, Science Education and Research Center (SERC)
Tel: (215)204-2071, Fax: (215)204-5082
xubin dot he at Temple dot edu
Dr. Xubin He's research has been mainly sponsored by the U.S. National Science Foundation (NSF). He has authored and co-authored many technical
articles in prestigious journals such as IEEE Transactions on Parallel and Distributed Systems (TPDS), Journal of Parallel and
Distributed Computing (JPDC), ACM Transactions on Storage (ToS), and IEEE Transactions on Dependable and Secure
Computing (TDSC), and at international conferences, including Conference on File and Storage Technologies (FAST), EuroSys, USENIX Annual Technical Conference (ATC), ACM International Conference on Supercomputing (ICS), IEEE/IFIP International Conference on Dependable Systems and Networks (DSN), IEEE International Parallel and Distributed Processing Symposium (IPDPS), IEEE International Conference on Local Area Networks (LCN),
etc. Dr. He has been conducting extensive research in the area of data storage and I/O systems. His research interests include networked storage, big data management, storage cache and disk I/O, high availability and reliable computing. Graduate students who have graduated under his supervision usually work in computer industry or research institutions as computer engineers or design architects. Dr. He has served as a panelist for NSF, also as a general/program chair and a technical committee member for many professional conferences in the field. He received the Ralph E. Powe Junior Faculty
Enhancement Award in 2004 and the Sigma Xi Research Award (TTU Chapter) in 2010 and 2005. He is a senior member of the IEEE, a member of the IEEE Computer Society, ACM, and USENIX. Prior to joining Temple, he was a faculty member at VCU (2010-2016) and Tennessee Tech University (2002-2010).
Selected Publications (Full List)
- [USENIX FAST: USENIX Conference on File and Storage Technologies]
- "GearDB: A GC-free Key-Value Store on HM-SMR Drives with Gear Compaction", Boston, Feb. 24-28, 2019 (acceptance rate: 26/145=18%).
- "Design Tradeoffs for Data Deduplication Performance in Backup Workloads", Santa Clara, CA, Feb. 16-19, 2015 (acceptance rate: 28/130=21.5%).
- "Reducing SSD Read Latency via NAND Flash Program and Erase Suspension", San Jose, CA, Feb. 14-17, 2012 (acceptance rate: 26/137=19%).
- [USENIX ATC: USENIX Annual Technical Conference]
- "MatrixKV: Reducing Write Stalls in LSM-based KV-stores with a Matrix Container in NVM", Boston, July 15-17, 2020 (acceptance rate: 65/348=18.6%).
- "RAFI: Risk-Aware Failure Identification to Improve the RAS in Erasure-coded Data Centers", Boston, July 11-13, 2018 (acceptance rate: 76/378=20%).
- "FlexECC: Partially Relaxing ECC of MLC SSD for Better Cache Performance", Philadelphia, June 17-20, 2014 (acceptance rate: 36/241=15%).
- "Accelerating Restore and Garbage Collection in Deduplication-based Backup Systems via Exploiting Historical Information", Philadelphia, June 17-20, 2014 (acceptance rate: 36/241=15%).
- "Victim Disk First: An Asymmetric Cache to Boost the Performance of Disk Arrays under Faulty Conditions", Portland, OR, June 15-17, 2011 (acceptance rate: 27/180=15%).
- [EuroSys: The European Conference on Computer Systems]
- "An Efficient Page-level FTL to Optimize Address Translation in Flash Memory", France, April 20-24, 2015 (acceptance rate: 32/150=21%)
- "An Aggressive Worn-out Flash Block Management Scheme to Alleviate SSD Performance Degradation", Amsterdam, The Netherlands, April 13-16, 2014 (acceptance rate: 27/147=18%).
- "Delta FTL: Improving SSD Lifetime via Exploiting Content Locality", Bern, Switzerland, April 10-13, 2012 (acceptance rate: 27/178=15%).
- [DSN: IEEE/IFIP International Conference on Dependable Systems and Network]
- "Power-aware Checkpointing: Toward the Optimal Checkpointing Interval under Power Capping", Toulouse, France, June 28-July 1, 2016 (acceptance rate: 58/259=22%).
- "HDP Code: A Horizontal-Diagonal Parity Code to Optimize I/O Load Balancing in RAID-6," Hongkong, China, June 27-June 30, 2011 (acceptance rate: 26/148=17.6%).
- "Code-M: A Non-MDS Erasure Code Scheme to Support Fast Recovery from up to Two-Disk Failures in Storage Systems", Chicago, June 28-July 1, 2010 (acceptance rate: 40/174=23%).
- [IPDPS: IEEE International Parallel & Distributed Processing Symposium]
- "StragglerHelper: Alleviating Straggling in Computing Clusters via Sharing Memory Access Patterns", New Orleans, May 18-22, 2020
- "EC-Fusion: An Efficient Hybrid Erasure Coding Framework to Improve Both Application and Recovery Performance in Cloud Storage Systems", New Orleans, May 18-22, 2020
- "Optimizing the Parity Check Matrix for Efficient Decoding of RS-based Cloud Storage Systems”, Rio de Janeiro, Brazil, May 20-24, 2019
- "Understanding and Modeling Lossy Compression Schemes on HPC Scientific Data", Vancouver, May 21-25, 2018 (acceptance rate: 24.5%), Best paper nominee.
- "A Set-aware Key-Value Store on Shingled Magnetic Recording Drives with Dynamic Band", Vancouver, May 21-25, 2018 (acceptance rate: 24.5%).
- "Chameleon: An Adaptive Wear Balancer for Flash Clusters", Vancouver, May 21-25, 2018 (acceptance rate: 24.5%).
- [ICS: ACM International Conference on Supercomputing]
- "Demystifying Cache Policies for Photo Stores at Scale: A Tencent Case Study", Beijing, China, June 2018 (acceptance rate: 36/193=18.7%).
- "A Stall-Aware Warp Scheduling for Dynamically Optimizing Thread-level Parallelism in GPGPUs", Newport Beach, CA, June 2015 (acceptance rate: 40/160=25%).