University of Minnesota D Digital Technology Center I Intelligent S Storage C Consortium

1
University of Minnesota D Digital Technology
CenterI Intelligent S Storage C Consortium
2
Overall Focus

• Emphasize the application of Advanced Storage
  Technologies
• A Balanced approach to research that includes
• Applications that need/use storage
• Advanced and Emerging Storage Architectures
• Advanced and Emerging Storage Technologies both
  software and hardware
• Business Cases and aspects of the Storage
  industry
• Market Trends
• Product Directions
• Effects of these disruptive technologies
• Adoption rates
• Provide consortium members with not just
  technology research but a more complete and
  significant outcome

3
Initial Specific Focus

• Applied Object-based Storage Device (OSD) Active
  Storage
• The Application of OSD and Active Storage Devices
  to different real-world problems
• Demonstrate the feasibility and advantages of OSD
  and Active Storage
• Develop an understanding of the limitations of
  OSD and Active Storage from a theoretical and
  practical standpoint
• How OSD Active Storage addresses real-world
  problems facing the storage industry today and
  tomorrow
• Must make this a Win-Win value proposition for
  the University and Industry

4
University Participation

• The Digital Technology Center
• Jim Licari and Tom Ruwart
• Cross-disciplinary University participants
• Computer Science Software Technology focus
• David Du Active disk, OSD, and Networking
• Jon Weissman Software systems, grid computing
• Yongdae Kim- Computer and Network Security
• Zhili Zhang Networking and Internet Engineering
• Electrical Engineering Hardware Technology
  focus
• Ahmed Tewfik Signal processing, wireless
  network
• David Lilja Computer Architecture, Distributed
  systems
• Carlson School of Business Business focus
• Bob Kauffman, Information and Decision Sciences
• Alok Gupta
• Gediminas Adomavicius

5
Related Links

• Universities
• University of California Santa Cruz Storage
  Systems Research Center - http//ssrc.cse.ucsc.edu
  /
• Carnegie Mellon University (CMU) Parallel Data
  Lab (PDL) http//pdl.cmu.edu
• University of California San Diego Information
  Storage industry Center - http//isic.ucsd.edu
• National Research Centers
• National Center for Supercomputer Applications
  (NCSA) www.ncsa.uiuc.edu
• San Diego Supercomputer Center (SDSC)
  www.sdsc.edu
• National Labs
• DoE
• Los Alamos National Labs www.lanl.gov
• Lawrence Livermore National Labs www.llnl.gov
• Sandia National Labs www.sandia.gov
• Fermi National Accelerator Lab www.fnal.gov
• DoD
• Army High Peformance Computing Research Center
  www.ahpcrc.umn.edu
• Naval Research Lab www.nrl.gov
• Scientific Organizations
• NASA - www.nasa.gov
• National Radio Astronomy Observatory (NRAO)
  www.nrao.edu

6
Whats in it for the University

• Provides ample supply of focused, real-world
  projects and funding for Masters and PhD Thesis
  work
• Connects students more closely with industry
  giving them a more complete education
• Technology transfer from the University to
  Industry and vice versa (possibility of shared
  Intellectual Property)
• Research projects with industrial partners can be
  leveraged to obtain grants for research in other
  areas
• The DTC can act as a catalyst to bring other
  colleges and departments together to focus
  research on real-world interdisciplinary problems
  and issues

7
Whats in it for the Industrial Partners

• Ability to participate in more complete and
  significant research that is Application-driven
  and incorporates Storage Architectures,
  Technologies, and Relevant Business issues
• Use of research staff and students to explore and
  study real-world problems and issues
• Access to a pool of well-trained engineers for
  hiring
• Access to research facilities and cost-effective
  students
• Assistance in developing proof-of-concept
  technology demonstrations involving real-world
  problems and issues
• A common ground to meet and work with other
  companies on pre-competitive problems and issues
  related to the storage industry at large
• Funding leverage with other federal and state
  funding

8
Active Data Object based on Intelligent Storage
Concept

• Intelligent Storage Consortium
• David H.C. Du

9
What is happening?

• Computing devices with large storage capacity
  becoming pervasive
• Wireless and mobile devices becoming popular
• Storage systems becoming cheaper and larger
• The volume of available data becoming extremely
  large and hard to manage

10
Future Computing Environment

• Global Internet reach everywhere
• Pervasive Computing include many appliances with
  wireless ad hoc networks
• Intermittent Connectivity
• Large storage capacity in each device
• Data duplication is a must
• Dynamically changed user demand

11
Our Initial Focus

• Propose and develop the active data object
  concept
• Based and extending the OSD (Object Storage
  Device) Standards
• Apply the OSD and intelligent storage concept to
  future computing environment
• Investigate applications and environments that
  can benefit directly

12
Object Storage Model

• An object is a logical unit of storage
• Lives in flat name space with an ID
• Contains data and metadata (similar to an inode)
• File-like methods open, close, read, write
• An OSD stores objects and could be any of
• Disk drive, storage appliance, storage
  controller,
• OSDs enables high performance and cross platform
• Use the higher level abstraction we needed
• Offload read write from the storage server

Objects enable self-managed storage.
13
OSD System Architecture
I/O Application
Storage System
Manager
OPEN/CLOSE
Storage Device
READ/WRITE
MANAGEMENT
The Manager is not in the data path.
14
Proposed Extension to OSD

• Data Ownership
• Data Access Rights
• Data Access Log
• Data Encryption Information
• Current Version Number
• Meta Data Manager Location
• Expected I/O Performance Requirement
• Potential Data Processing Methods

15
Current Status

• Research projects are defined
• 5 faculty and 6 research assistants (supported by
  DTC seed funding) are involved
• Actively soliciting industrial partners
• Aggressively seeking federal funding

16
David H.C. Du

• Academic Experience
• With Department of Computer Science and
  Engineering, University of Minnesota since 1981
• IEEE Fellow since 1998
• Industrial Experience
• Senior Consulting, ITRI/CCL Taiwan, 1996
• VP of Engineering at 3CX, 1998 lead a team of
  30 engineers working on ATM switches, ATM NICs,
  Fast Ethernet switches, and streaming video
  severs
• Chairman CEO, Streaming21, 2001 Raised 18M
  and focus on streaming video software products

17
David Dus Current and Past Research Interests

• 1980s physical database design,
  parallel/distributed processing
• 1990s CAD for VLSI circuits, computer
  networking
• 2000s multimedia computing, high-speed and
  optical networks, mass storage systems
• Published more than 150 papers including 75
  journal articles Graduated 37 Ph.D. and 65 M.S.
  Students

18
David J. Lilja

• Academic experience
• Professor, Electrical and Computer Engineering,
• University of Minnesota (1991-present)
• Director of Graduate Studies, Computer Engr
  (1996-1998)
• Visiting Professor, University of Western
  Australia (2001)
• Ph.D., Electrical Engineering,
• University of Illinois at Urbana-Champaign
  (1991)
• Industrial experience
• Visiting senior engineer, Future Processor
  Performance Group,
• IBM, Rochester, Minnesota (2000)
• Processor development engineer,
• Tandem Computers, Inc., Cupertino, California
  (1982-1986)

19
David J. Lilja

• Research Interests
• High-performance computer architecture
• Parallel and distributed systems
• Multiprocessor memory/storage hierarchies
• Performance measurement and analysis
• Compilers
• Hardware/software co-design
• Molecular computing (nanocomputing)

20
Zhi-Li Zhang

• Ph.D, U. of Massachusetts, Feb 1997
• Assistant Professor, Jan 1997-May 2002
• McKnight Land-Grant Professorship, 2000-2003
• Associate Professor, Fall 2002
• Visiting positions at many industrial RD labs
• Sprint ATL, Fujitsu Labs, IBM T.J. Waston,
• Editors for
• IEEE/ACM Transactions on Networking
• International Journal of Computer Networks
• Served on many conference/workshop committees

21
Zhi-Li Zhang Research Interests

• Networking and Internet Technology
• performance and quality of service
• routing and network engineering
• Multimedia Systems
• video streaming techniques
• Networked Storage Systems
• network support and qualify of service
• Distributed Computing Systems
• grid computing, service discovery and routing

22

• Jon B. Weissman Department of Computer Science
• University of Minnesota
• http//dcsg.cs.umn.edu
• jon_at_cs.umn.edu

Education and Background Ph.D. UVa 1995 (Grimshaw
advisor) B.S. CMU 1984 Industry experience Mitre
Corp 1989-1991 Key architect of Legion project at
U of Va
23

• Research interests are in distributed systems,
  high-performance computing, Grid computing,
  distributed storage.
• Has published 40 referred papers in these areas
• Leads Distributed Computing Systems Group 10 in
  group
• Most recent project is community services
• Developed several scheduling systems Prophet,
  Gallop
• Current research is funded by NSF, AHPCRC, DTC,
  and other sources

24
Yongdae Kim

• Assistant Professor, CS of UMN
• Education
• Ph.D. USC, 2002
• Advisor Gene Tsudik
• Thesis Group Key Agreement theory and practice
• MS,BS Yonsei Univ, Korea 1993, 1991
• Emplyment History
• Jan. 2000 Jun. 2002 UC Irvine, visiting
  researcher
• Sep. 1998 Dec. 2000 USC/ISI, research
  assistant
• Feb. 1993 Jun. 1998 ETRI, Korea, research
  staff
• URL http//www.cs.umn.edu/kyd.

25

• Research Interests
• Network Security
• Peer Group Key Agreement, Group Signatures,
  Public Key Infrastructures, Access Control
• Multicast Security Key Distribution, Stream
  Authentication
• Anonymous Communication
• Denial of Service Attacks Prevention and
  Recovery
• Security of Mobile Ad Hoc Networks
• Storage Area Network security
• Distributed Systems
• Fault-tolerant group communication
• Peer-to-peer Systems file sharing, content
  distribution
• Cryptography Random Number Generators, Digital
  Signatures, Block and Stream Ciphers

26
Ahmed Tewfik

• E. F. Johnson Professor of Electronic
  Communications
• BS in Electrical Engineering from Cairo
  University
• ScD and Master in Science degrees in EE and
  Computer Science from MIT
• Awards and Honors- IEEE 3rd Millennium Award,
  Fellow of the IEEE, Distinguished lecturer of the
  IEEE, George Taylor Faculty Award, NSF Research
  initiation award, Plenary speaker at numerous
  IEEE conferences
• Founder CEO of Cognicity, sold to Digimarc
• Founded and led IEEE publication, IEEE Signal
  Processor letters
• Consultant to Emerson-Rosemount and MTS
• Areas of research High speed wireless networking
  for storage and multimedia production/post-product
  ion/distribution, data centric computing and
  communications, I/O for storage area networks
  and parallel computing, multimedia analysis,
  retrieval and protection, heart diagnostics