Xiongzi Ge

   

Xiongzi Ge      葛雄资


NetApp
Office: Morrisville, NC 27709
Contact: xiongzi at cs dot umn dot edu

| Biography | Education | Research| Work/Intern Experience | Publications |Patents |


Biography

I joined NetApp as a member of the Advanced Technology Group in May 2016. Prior to joining
NetApp, I was a member of the Center for Research in Intelligent Storage (CRIS) at University of Minnesota from 2013 to 2016. I successfully defended my PhD thesis in 2017 at department of computer science and engineering, University of Minnesota - Twin Cities. My advisor is Prof. David H.C. Du. I received the doctor degree in Computer System Architecture, advised by Prof.Dan Feng , in Data Storage and Application Lab, from Huazhong University of Science and Technology (HUST), Wuhan, China, in 2012. I worked as a software engineer and was leading a small data storage group in Shenzhen Cloud Computing Center (National Supercomputing Center in Shenzhen) from 2011 to 2013. Before that, I was a visiting scholar in the Shenzhen Institutes of Advanced Technology(SIAT), Chinese Academy of Sciences, from 2009 to 2011. I was a visiting student in the Digital Technology Center(DTC), in University of Minnesota, from 2008 to 2009.


Education

Ph.D., Computer Science, University of Minnesota, Twin Cities, 2017
M.S., Computer Science, University of Minnesota, Twin Cities, 2016
Ph.D., Computer System Architecture, Huazhong University of Science and Technology, Wuhan, China, 2012
B.S., Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China, 2005


Research
Major projects participated in U of M:

ChewAnalyzer: Chunk Level Data Management across Differentiated Storage Pools
This work designs a storage-aware workload profiling method, simplified as ChewAnalyzer, to consolidate block-level heterogeneous storage systems.

HintStor: A Flexible Framework to Study I/O Access Hints in Heterogeneous Storage Systems
In this project, in the beginning, we aimed to seek an efficient way to integrate local and off-premise cloud storage. We did some study to incorporate local storage with OpenStack cinder block storage using access hints. We exploited file system and application level access hints for block level tiered storage management in current Linux. We develop a prototype, called HintStor, in Ubuntu 14.04 (kernel 3.13.0), which includes a framework to evaluate access hints based research. HintStor can execute and evaluate various I/O access hints on heterogeneous storage systems with minor modifications to the kernel and applications. The design of HintStor contains a new application/user level interface, a file system plugin and a block storage data manager. With HintStor, storage systems composed of various storage devices can perform pre-devised data placement, space reallocation and data migration polices assisted by the added access hints.

OpenANFV: Accelerating Network Function Virtualization
To consolidate various hardware resources in an elastic, programmable and reconfigurable manner, we design and implement a flexible and consolidated framework, OpenANFV, to support acceleration for multiple vi
rtualized Middleboxs (Dedup, DPI, NAT, etc.) in the cloud environment. For instance, we designed and developed an online appliance for network redundancy elimination (NRE) using software based control and FPGA based acceleration. This dedicated appliance uses a virtualized controller called VNRE, to deal with multiple TCP flows with high performance and low cost flexibly.

My research interests include file systems, block storage and other topics about big data research.


Work/Intern Experience

Summer Research Assistant, Advanced Technology Group, NetApp, 2015.05-2015.08
Designed and developed Garbage Collection policies for host-managed Shingled Magnetic Recording disks.

Summer Research Assistant, NEC Laboratories America, Inc, 2014.05-2014.08
Migration-Aware Server Side SSD caching in QEMU. Let QEMU support efficient SSD caching after Virtual Machine Live Migration.

Software Engineer, R & D department, National Supercomputing Center in Shenzhen (Shenzhen Cloud Computing Center), 2011-2013
In NSCCSZ, I worked as a consultant of storage research and development. I was leading a 8-people storage group to develop cloud computing and cloud storage application software;participated the development of PengYun Project about Cloud Storage and Services; participated the deployment of a cloud backup project and Openstack project


Publications

Conference papers:


Peter Macko, Xiongzi Ge, John Haskins Jr., James Kelley, David Slik, Keith A. Smith and Maxim G. Smith. SMORE: A Cold Data Object Store for SMR Drives. In 33rd International Conference on Massive Storage Systems and Technology, Santa Clara, CA, 2017 (MSST 2017).

Fenggang Wu, Ming-Chang Yang, Ziqi Fan, Baoquan Zhang, Xiongzi Ge, David H.C. Du. Evaluating Host-Aware SMR (HA-SMR) Drives. In the 8th USENIX Workshop on Hot Topics in Storage and File Systems, Denver, CO, 2016 (HotStorage'16).

Xiongzi Ge, Yi Liu, Chengtao Lu, Jim Diehl, David H.C. Du, Liang Zhang and Jian Chen. VNRE: Flexible and Efficient Acceleration for Network Redundancy Elimination. In the 30th IEEE International Parallel & Distributed Processing Symposium, Chicago, IL, 2016 (IPDPS 2016).

Yi Liu, Xiongzi Ge, David H.C. Du and Xiaoxia Huang. Par-BF: a Parallel Partitioned Bloom Filter for Dynamic Data Sets. In the workshop on Data-Intensive Scalable Computing Systems (DISCS), in conjunction with SC'14. November 16th, 2014, New Orleans LA, USA.

Yi Liu, Xiongzi Ge, Xiaoxia Huang, David H.C. Du. MOLAR: A High Performance, Cost-Efficient Hybrid Storage Cache. In the IEEE Cluster 2013 Conference (Cluster'13), September 23-27, 2013, Indianapolis, USA.

Yujuan Tan, Zhichao Yan, Dan Feng, Edwin Hsing-Mean Sha, Xiongzi Ge. Reducing the De-linearization of Data Placement to Improve Deduplication Performance. In the workshop on Data-Intensive Scalable Computing Systems (DISCS), in conjunction with the 2012 ACM/IEEE Supercomputing Conference (SC'12), Saltlake city, Utah, USA, 2012.

Jiao Hui, Xiongzi Ge, Xiaoxia Huang, Yi Liu, Qiangjun Ran. E-HASH: An Energy-Efficient Hybrid Storage System Composed of One SSD and Multiple HDDs. In the Third international conference on Advances in Swarm Intelligence (ICSI'12),LNCS, 2012, Shenzhen, China.

Xiongzi Ge, Dan Feng, David H. C. Du. DiscPOP: Power-aware buffer management for disk accesses. In the Proceedings of 2011 IEEE International Green Computing Conference and Workshops (IGCC'11), July 25-28, Orlando, USA.

Chao Jin, Dan Feng, Hong Jiang, Lei Tian, Jingning Liu, Xiongzi Ge. TRIP: Temporal Redundancy Integrated Performance Booster for Parity-Based RAID Storage Systems. In the Proceedings of the 16th International Conference on Parallel and Distributed Systems (ICPADS'10), Shanghai, China, December 2010.(Best Paper Award Nominee)

Xiongzi Ge, Dan Feng, Chengtao Lu, Chao Jin. Dynamics Analysis Model for Energy Efficient Distributed Storage Systems. In International Conference on Environmental and Computer Science (ICECS), 2010, Kunming, China.

Xiongzi Ge, Dan Feng, Lei Tian. An Active Region-based Storage Mechanism in Large Wireless Sensor Networks. In the Proceedings of 2007 International Conference on Networking, Architecture, and Storage (NAS'07), Guilin, China, July 2007.

Journal papers:

Fenggang Wu, Ming-Chang Yang, Ziqi Fan, Baoquan Zhang, Xiongzi Ge, David H.C. Du. Performance Evaluation of Host Aware Shingled Magnetic Recording (HA-SMR) Drives. In the IEEE Transactions on Computers, 2017.

Dingde Jiang, Yang Han, Zhihan Lv, Lei Miao, Ting Zhu, and Xiongzi Ge. Dynamic access approach to multiple channels in pervasive wireless multimedia communications for technology enhanced learning." Journal of Intelligent & Fuzzy Systems, 2016.

Dingde Jiang, Lei Shi, Peng Zhang, Xiongzi Ge. QoS constraints-based energy-efficient model in cloud computing networks for multimedia clinical issues. Multimedia Tools and Applications, 2016.

Yi Liu, Xiongzi Ge, David H.C. Du and Xiaoxia Huang. Par-BF: a Parallel Partitioned Bloom Filter for Dynamic Data Sets. In the International Journal of High Performance Computing, 2015.

Yi Liu, Xiongzi Ge, David H.C. Du and Xiaoxia Huang. SSD as a Cloud Cache? Carefully Design about It. In Journal of Computers, 2015.

Yi Liu, Xiongzi Ge, Xiaoxia Huang, David H.C. Du. MOLAR: A High Performance, Cost-Efficient SSD-based Hybrid Storage Cache. In the Computer Journal, 2015. (pdf)

Xiongzi Ge, Dan Feng and David H.C. Du. DiscPOP: Power Aware Buffer Management for Disk Accesses. In Journal on Sustainable computing: Informatics and Systems (SUSCOM), Elsevier, Volume 3, Issue2, July,2013.

Ge Xiong-zi, Dan Feng, Chengtao Lu, Chao Jin. Dynamics Analysis Model of Green Network Storage Systems. In Journal of Computer Science (in Chinese), in August, 2011.

Xiongzi Ge. The Dynamics Rules of Complex Network Storage System. Bulletin of Advanced Technology Research , in SIAT, April, 2010.

Lu Cheng-tao, Feng Dan, Wang Fang, Ge Xiong-zi. A Statistical Analysis-Based Approach for Storage System Performance Tuning. In Journal of Computer Science, November, 2010 (in Chinese).

Chen Jin-Cai, He Pin, Ge Xiong-zi. A Dynamics Analysis Method of Cellular Automata for Complex Network Storage System. Journal of Software, Oct, 2008.

Posters/Demos/Work-in-Progress:

Xiongzi Ge, Zhichao Cao, Pradeep Ganesan, David H.C. Du, and Dennis Hahn. OneStore: Integrating Local and Cloud Storage with Access Hints. In the Poster Session of 2014 ACM Symposium on Cloud Computing. November 3-5, 2014, Seattle, WA.

Xiongzi Ge, Yi Liu, David H.C. Du, Liang Zhang, Hongguang Guan, Jian Chen, Yuping Zhao and Xinyu Hu. OpenANFV: Accelerating Network Function Virtualization with a Consolidated Framework in OpenStack. In Proceedings of the 2014 ACM conference on SIGCOMM, August, 2014, Chicago, USA.(A demo in SIGCOMM'14)


Patents

Applied over 11 Patents (6 granted) and software copyrights in data storage and cloud computing field.


Homepage Visitors