Krishna Kant
Office phone: (703)292-4776, (703) 993-5152
Email: kkant@nsf.gov, kkant@gmu.edu
Home page : http://www.kkant.net
Education:
Ph.D. in Computer Science, University of Texas at
Dallas, Richardson, TX (1981).
M.S. in Electrical Communication Eng., Indian
Institute of Science, Bangalore, India (1977).
B.Tech. in Electrical Eng., Indian Institute of
Technology, Kanpur, India (1975).
Current Research Interests: Security and robustness issues in cloud
computing and name resolution, Configuration management issues in systems,
Energy Efficiency and Sustainability issues in computing & communications,
emerging nonvolatile memory technologies.
Other expertise: Traffic characterization and congestion
control, Low-level architectural modeling, Acceleration technologies (security,
XML, TCP, etc.), Peer to Peer computing, Mathematical performance modeling,
Telecommunication systems and Internet technologies.
Professional Experience: Assistant and then tenured associate professor
of computer science (1981-1991), R&D in telecommunications industry
(1991-1997), R&D in computer industry (1997-2008), NSF program director
(2008 - April 2010), Visiting professor (April 2010 - present), NSF program
director (Sept 2010 - present).
Books and Book Chapters Published:
Recent
Funded Grants:
Note: In grants 1-3, my role is listed
as senior personnel because of my industry affiliation; however, I was the
primary intellectual contributor in all of these grants.
Recent Ph.D.’s co-advised
1.
Amit Sahoo (2008), UC/Davis, Now with Cisco Systems,
Thesis area: BGP Convergence under large scale internet failures
2.
Lihua Yuan (2009), UC/Davis, Now with Microsoft,
Thesis area: Cooperative Security Mechanisms for Domain Name System
3.
Neha Udar (2009), SIU, Now with Intel Corp., Thesis
area: Asset Localization in data centers using UWB radios
4.
Casey Deccio (2011), UC/Davis, Now with Sandia
National Labs, Thesis area: Enhancing Security of the Domain Name System
5.
Gim Jongmin (2011), Hanyang University, South Korea,
Post doc, Thesis area: Modeling emerging storage systems
6.
Meixing Le (in progress), George Mason University,
Thesis area: Secure collaboration in multi-cloud environments
7.
Muthu Murugan (in progress), University of
Minnesota, Thesis area: Energy Adaptation in Data Centers
8.
Mayank Raj (in progress), University of Texas at
Arlington, Thesis area: Energy adaptation for Mobile devices
Patents:
An Error Monitoring Algorithm for Broadband Signaling Networks, Oct 1996, US Patent no 5487072. US Patent also received for a variant of this algorithm, Patent no 5563874, US patent no 7027460 on peer to peer TV, and US patent no 7885914 coordinated memory rank power control. Patents in progress: Proactive link power control algorithm, coupled resource power control,
Editorial board:
Computer
Networks (COMNET), Intl journal of communications systems (IJCS), Sustainable
Computing Journal (SUSCOM), Electronic Commerce Research Journal.
Recent Tutorials:
Design
of Sustainable Data Centers (HPCA 2012, Intl. supercomputing conf. 2011),
Power/Thermal Challenges in Data Centers (Super-Computing 2009), Surviving
Large Scale Internet Failures (DSN 2007 and ICDCN 2009), Peer to Peer Computing
(Globecom 2003).
Tools developed :
1. LMPOWER, a comprehensive tool for detailed
studies on power management of memory and interconnect in SMP systems.
2. Geist (Generator of e-commerce and internet
traffic) available at http://www.kkant.net/geist
3. SimP2 (Tool for evaluating performance of
peer-to-peer file-sharing networks).
4. DCLUE (Database cluster emulator) as tool for
studying database cluster performance. Available at http://www.kkant.net/DCLUE
Recent program committees:
Infocom 2012-2010, ICCCN 2012, ICPP 2012, CCGrid
2011-12, ICDCS 2010, WASA 2010, ICDCN 2012-2009, ICPP 2010, ICPP 2009.
Employment History
2010 - Present: Research professor, Center for Secure Information
Systems (CSIS), George Mason University.
Currently also serving as a Program director in the CISE/CNS division at
the National Science Foundation
At GMU, I conduct research in two major areas: security and robustness issues in data centers and cloud computing infrastructure, and energy efficient/sustainable computing in the context of data centers. At NSF, I manage computer systems research (CSR) program in the CNS division and actively support a few other programs such as Expeditions in Computing and Cyber-physical systems. I am also deeply involved in the development of solicitations for the NSF-wide initiative called SEES (Science, Engineering and Education for Sustainability) and represent CISE directorate in this initiative at multiple levels. I have also been actively driving international collaboration initiatives within NSF in emerging fields of pervasive computing, sustainability, and infrastructure security. Also co-advised four Ph.D. students, two of whom have already graduated.
2008 - 2010: Program director, CISE/CNS division, National Science Foundation
A position in a similar capacity as the one above. In addition to managing CSR and other NSF programs, I started an initiative on energy efficiency of IT, particularly data centers. This sub-area within CSR continues to attract increasing interest from the community. I also started initiative in enhanced international collaboration and organized several workshops for international collaboration targeted at areas like pervasive computing & social networking, infrastructure security, and computer networks and distributed systems. Co-advising of 3 Ph.D. students.
Current Research Collaborations: I have active collaborations with several
universities. I have collaborated closely with University of California
at Davis since 2004 in the areas of robustness issues in the Internet, with
University of Minnesota on energy efficient and sustainable computing since
2009, with University of Texas at Arlington on energy adaptation in peer to
peer networks since 2009, and with Hanyang University (Korea) on emerging
nonvolatile memory (NVRAM) technologies since 2008.
1997 - 2008: Senior Performance Analysis Engineer, Digital Enterprise
Group, Intel Corp, Hillsboro, OR.
Mainstream R&D Areas: A significant part of my work was focused in designing new algorithms for power control of platform resources and detailed evaluation of their performance impact. Another key research was in the area of asset localization in data centers using wireless USB radios and its platform implementation issues. My related work spanned various aspects of data centers including real-time resource management, network virtualization, high speed networking, advanced transport layer features, metering and monitoring in a virtualized environments, hardware protocol acceleration, etc.
Other work: One major theme of my work in Intel was the detailed architectural modeling of server platforms from a variety of perspectives. These models typically include details of processor busses/links, memory pipeline, IO busses, network processing, etc. and were used to assess impact of a variety of new HW capabilities on the performance of server benchmarks such as TPC-C, TPC-H, SpecWeb 99/2005, etc. These models were routinely used in deciding overall architecture of Intel servers for various segments. In particular, I extensively studied hardware acceleration of various sorts (e.g., TCP, XML, encryption, authentication, compression, etc.), and the impact of inter-process communication latency on the performance of clustered database systems. I also worked on a variety of other research/implementation projects at Intel, including new server architectures that integrate compression into the DRAM, Internet server traffic characterization in terms of self-similarity and transactional properties, peer to peer computing, and network virtualization and utility computing.
Research Collaboration and Funding: During my tenure at Intel, I
collaborated extensively with UC/Davis and wrote several joint NSF funding
proposals. Two of these proposals were funded and supported much of my
joint research with UC/Davis. In addition, during 2004-5, I spent one
year as a visiting researcher at UC/Davis to study advanced transport features
for data center networks. This visit was entirely supported by a
competitive research grant funded by Intel Research. During this time I
also initiated a new project on meta-data channel based security in the
Internet and was successful in having it funded (along with UC/Davis) by Intel
IT-research (different from Intel research). The work on these projects
involved close supervision of two Ph.D. students. I also worked closely
with Southern Illinois University and advised a Ph.D. student there on the
topic of UWB radio based asset localization in data centers.
1991-1997: Senior Member of Technical Staff; Engineering,
Performance, and Control department, Telcordia (formerly Bellcore),
Red Bank, NJ.
R&D Areas: Telecommunications systems cover two major areas of R&D: operations support system (OSS) and switching systems (SS), and I have extensive experience on both of these sides. On the OSS side, I examined issues relating to automatic provisioning of telephone service which requires interactions with dozens of database systems. I developed both mathematical and simulation models to study completion times for automated provisioning. On the switching side, I examined a variety of issues relating to signaling associated with call management. In particular, I worked extensively on congestion control and link error monitoring in both narrow-band and broadband SS7 (signaling system No. 7) networks. I also examined a variety of engineering issues relating to SS7, personal communication systems (PCS), ISDN, AIN (advanced intelligent network), impact of data connections on voice networks, etc. Much of this work was done in support of Bellcore’s generic requirements for US regional telecommunications networks and formed a part of those requirements. During this time, I published extensively in the areas of congestion control, performance modeling, and error monitoring.
1985-1990: Assistant (1985-1989) and then tenured associate professor of Computer Science at Pennsylvania State University, University Park, PA.
Teaching, Research and Funding: I taught undergraduate courses in operating systems
and database systems which routinely enjoyed overflow enrollments and very high
student scores. I also developed and taught graduate courses in several
advanced areas including computer system performance modeling, advanced operating
systems, distributed systems, and fault-tolerant computing. My primary areas of
research were computer system performance modeling and fault tolerant
computing, and the research was funded by NSF and DoD. During this time,
I also wrote a highly regarded graduate textbook on mathematical performance
modeling (listed above). I personally established, managed, and supported
the first networked laboratory of SUN workstations at Penn State for research
and teaching in networked systems.
1981-1984: Assistant Professor of Electrical Engineering and Computer
Science, Northwestern University, Evanston, IL.
Teaching/Research: Helped develop and taught undergraduate courses in digital
circuits, compiler design, and operating systems. Designed and delivered
graduate courses/seminars in fault-tolerant computing, operating systems, and
software engineering. Carried out research in fault-tolerant software and
related areas.
External Publications (Since 1985):
1. K. Kant and A. Silberschatz, ``Error Propagation and Recovery in Concurrent Environments'', The Computer Journal, Vol 28, Nov 1985, pp466-473.
2. K. Kant, ``Finding Interference between Rectangular Paths'', IEEE Trans. on Computers, Vol C-34, Nov 1985, pp 1045-1049.
3. M.M Srinivasan and K. Kant, ``The File Allocation Problem: A Queuing Network Modeling Approach'', Computers and Operations Research, Vol 14, May 1987, pp349-361.
4. K. Kant, ``Software Fault-Tolerance in Real-Time Computing Environments'', Information Sciences, Vol 42, Aug 1987, pp255-282.
5. K. Kant, ``Performance Analysis of Hierarchical Ring Networks'', Proceedings of the 12th Conference on Local Area Networks,, pp95-108, Oct 1987.
6. K. Kant, ``Modeling Interprocess Communication in Distributed Programs'', Proc. of International Workshop on Petri-Net and Performance Models,, Madison, Wisconsin, pp. 75-83, Aug 1987.
7. K. Kant, ``Application Level Modeling of Parallel Machines'', Performance 88 Conference,, Santa Fe, NM , pp. 83-93, May 1988.
8. A. Ravichandran and K. Kant, ``Fault Identification in Robust Data Structures''. Proc. of 19th Fault Tolerant Computing Symposium,, Chicago, Illinois, pp. 275-282, June 1989.
9. M. Ghodsi and K. Kant, ``Performance Modeling of Concurrent Systems under Resource Constraints'', Parallel Computing 1989, pp 589-594, Elsevier Science publisher.
10. K. Kant and F.C. Liaw, ``Modeling Parallel Programs Using Dynamic Task Graphs'', Proc. of the first IEEE Symposium on Parallel and Distributed Processing., Dallas, Texas, pp. 90-97, May 1989.
11. K. Kant, ``Performance Analysis of Real-Time Software Supporting Fault-Tolerant Operation'', IEEE Transactions on Computers, 39(7), pp. 906-918, July 1990.
12. K. Kant, ``Analysis and Synthesis of Generalized Task Graphs'', Proc. of the 2nd Conf. on parallel and distributed systems, Dallas, Texas, pp. 225-231, Dec 1990.
13. M. Ghodsi and K. Kant, ``Analysis of Parallel Programs Running on Multiprocessor Systems'', Performance 90, Edinburgh, U.K., pp. 407-421, Dec 1990.
14. M. Ghodsi and K. Kant, ``Performance Analysis of Parallel Search Algorithms on Multiprocessor Systems'', Performance Evaluation, 13(1), pp. 67-83, sept 1991.
15. M. Ghodsi and K. Kant, ``Well-Formed Generalized Task Graphs'', Proc. of the 3rd symposium on parallel and distributed systems, Dallas, TX, pp. 344-351, Dec 1991.
16. K. Kant, ``Analysis of Delayed Repair Policies in Manufacturing Systems'', presented at ORSA-TIMS conference, Nashville, TN, May 1991.
17. K. Kant, ``MVA Approximation for Shortest-Job-Next Scheduling Discipline'', Performance Evaluation, Vol 14, March 1992.
18. K. Kant and A. Ravichandran, ``Synthesizing Robust Data Structures: An Introduction'', IEEE Transactions on Computers,, 29(2), pp. 161-173, Feb 1990.
19. K. Kant and A. Ravichandran, ``Synthesizing Robust Data Structures: Formal Approach'', Acta Informatica, 1992.
20. A. Ravichandran and K. Kant, ``A General Approach to the Identification Compensated Faults in Robust Data Structures'', Information Science, 59(1), pp. 167-187, Jan 1992.
21. D. Daly, K. Kant, Y-B. Lin, V. Mak, and D. Mok, ``COPS: A computer operations performance simulation system'', Proc of 26th Annual Simulation Symposium, Washington D.C., April 1993.
22. K. Kant, ``Steady State Analysis of Stochastic Systems'', Chapter 2, pp17-68, in Handbook of Statistics, Vol 9, 1993, edited by C. R. Rao.
23. K. Kant, ``Performance of Window Flow-Control Protocol under Errors'', April 1993, Bellcore (Telcordia) technical report.
24. K. Kant, ``Sustainable CCS Error Rate and its Implications for Error Monitoring'', April 1993, Bellcore (Telcordia) technical report.
25. K. Kant, ``Evaluation of Error Interval Monitoring Algorithm'', Sept 1993, Bellcore (Telcordia) technical report.
26. K. Kant, ``Error Monitor Design for High Speed CCS Links'', Sept 1993, Bellcore (Telcordia) technical report.
27. K. Kant, ``Estimating Transmit Congestion at Changeover Time in CCS Networks'', Sept 1993, Bellcore (Telcordia) technical report.
28. V.T. Hou, K. Kant, V. Ramaswami and J.L. Wang, ``Error Monitoring Issues for Common Channel Signalling'', IEEE Journal of Selected Areas in Communications, vol 12, no 3, April 1994, pp 456-467.
29. K. Kant, ``Performance of Internal Overload Controls in Large Switches'', 28th Annual Simulation Conference, Phoenix, AZ, April 1995, pp 228-237.
30. K. Kant, ``Broadband IN Signalling and Control Architectures and Comparison with Industry directions'', Bellcore TR-3745, Nov 1995.
31. K. Kant, ``Analysis of Delay Performance of ATM Signaling Links'', Proc of INFOCOM 95, Boston, MA, April 1995, pp 1146-1153.
32. ``Generic Requirements for CCS Nodes Supporting ATM High-Speed Signalling Links'', GR-2878-CORE, Bellcore, Oct 1995.
33. K. Kant, ``Analytic Modeling of SSCOP'', Proc of 3rd International conference on Telecommunication Systems, Nashville, TN, Feb 1995, 469-482.
34. K. Kant and J.R. Dobbins, ``An Error Monitoring Algorithm for ATM Signaling Links'', Proc of Sixth IFIP Conference on Performance of Computer Networks, Istanbul, Turkey, Oct 1995, pp 367-381.
35. K. Kant, ``A Simulation Study of BISUP Congestion Control Issues'', Dec 1995, Bellcore (Telcordia) technical report.
36. K. Kant, ``Flow Control Mechanisms for SAAL Links'', Proc of International IFIP-IEEE Conference on Broadband Communications, Montreal, Canada, April 1996, pp 173-184.
37. K. Kant, ``A Study of BISUP Call Processing Delay Objectives'', Proc of Globecom 96, London, UK, pp 1400-1404, Nov 1996.
38. K. Kant, ``A unified global congestion control strategy for broadband signalling networks'', Jan 1997, accepted for ATM’97 international conference.
39.
K. Kant, ``Flow control
issues in ATM signalling link deployment'', Proc. of 15th International
Teletraffic Congress, Washington DC, pp 1219-1228, June 1997.
41. K. Kant, ``Large Memory System Model for TPC-C Benchmark'', Sept 1997,
Intel technical report.
98. C. Deccio, J. Sedayao, K. Kant, and P. Mohapatra, “Quality of Name Resolution in the Domain Name System”, to appear in Elsevier Computer Networks Journal.