Home

Deprecated. Go here.

Joseph L. Hellerstein

Senior Data Science Fellow and Affiliate Professor

University of Washington

Seattle, Washington

business email: jlheller@uw.edu

personal email: joseph.hellerstein@gmail.com

Education

  • Ph.D. COMPUTER SCIENCE, University of California, Los Angeles, CA, 1984.
  • M.S. COMPUTER SCIENCE, University of California, Los Angeles, CA, 1979.
  • BA MATHEMATICS, University of Michigan, Ann Arbor, MI, 1974.

Selected Honors and Awards

  • IEEE Fellow for "contributions to control engineering of performance management of computing systems," November, 2009.
  • IFIP Stokesberry Award for “outstanding contributions to the network, system and service management community,” May, 2007. 
  • IBM Outstanding Innovation award (for contributions to control theory and its application to IBM’s DB2 database management product), 2006. 
  • IBM Master Inventor, 2005 for high value patents.
  • Best paper (“A Model of Configuration Complexity and Its Application to a Change Management System”), IEEE/IFIP Integrated Management, 2005.
  • Best paper (“Using MIMO Feedback Control to Enforce Policies for Interrelated Metrics With Application to the Apache Web Server”), IEEE/IFIP Network Operations and Management, 2002.

Professional Experience

    • Senior Data Science Fellow at the eScience Institute and Affiliate Professor in Computer Science and Bioengineering, University of Washington, Seattle, Washington (May, 2014 - present)
    • Manager, Computational Discovery, Google, Inc., Seattle, Washington (October, 2011 -- May, 2014).
    • Manager, Performance Analytics, Google, Inc., Seattle, Washington (October, 2008 – October, 2011).
    • Principal Architect, Microsoft Corp., Redmond, Washington (July, 2006 – September, 2008).
    • Senior Manager, IBM Research, Hawthorne, New York (September, 2001 - June, 2006.
    • Manager, IBM Research, Hawthorne, New York (September, 1998 - August, 2001).
    • Research Staff Member, IBM Research, Hawthorne, New York (April 1984 – August, 2001).

Teaching And Selected Invited Lectures        

  • Affiliate Professor, University of Washington, "Biochemistry for Computer Scientists" and Data Science Fellow.
  • Keynote, International Conference on Network and Service Management (and INRIA, Nancy, France), "Optimizing Google Compute Clusters," October 24, 2011.
  • Keynote, International Conference on Quantitative Evaluation of Systems. "The Role of Performance Models in Building Scalable Cloud Infrastructures," Williamsburg, VA, September 18, 2010.
  • Invited talk at Brown University Symposium on Cloud Computing, "The Role of Quantitative Models in Building Scalable Cloud Infrastructures," May 13, 2010.
  • Adjunct Professor, University of Washington at Seattle. Course title: Analysis and Control of Computing Systems Using Linear Discrete-Time System Theory. Winter, 2008.
  • Keynote, International Workshop on Self-Organization Systems, September, 2007.
  • Invited lecturer on application of control theory to computing systems at Lund and Linköping Universities, Sweden, November, 2007.
  • Distinguished Lecturer, Autonomic Computing Research Center, KTH, Stockholm, Sweden, May, 2007.
  • Adjunct Professor, Columbia University. Course title: Feedback Control of Computing Systems, Spring & Fall, 2004.
  • Adjunct Professor, Columbia University. Course Title: Theory and Application of Network and Systems Management, Spring, 1995 and Spring, 1996.

Recent Research Service Leadership Roles

    • Program committee co-chair, International Conference on Autonomic Computing, 2011.
    • Steering committee, Workshop on Feedback Control in Implementation and Design of Computing Systems, 2007-2009.
    • Co-organizer of First International Workshop on Feedback Control in Implementation and Design of Computing Systems, 2006.
    • Program Committee Co-Chair, IEEE/IFIP Network Operations and Management, 2006.
    • General Chair for the Third Systems Management Workshop, Newport, Rhode Island, April 22-24, 1998.
    • Editor, IEEE Transactions on Network and Systems Management, 2006 to present.
    • Board of directors of the Northeast Computer Measurement Group, 1997-2000.

Government Committees

    • DARPA Information Science and Technology (ISAT) advisory group, September, 2006-present.
    • Co-organizer NSF workshop on “Feedback Control of Computing Systems,” IBM Thomas J. Watson Research Center, Hawthorne, NY. May 3-4, 2005. (NSF Sponsor-Helen Gill)
    • Co-lead of DARPA workshop “Information Sciences and Technology, Benchmarks and metrics for cognitive systems”, May 4-5, 2004.
    • Co-organizer of NIST-ATP program "Operations and Management of Information Networks", 1993.

Major Projects

Big Science in the Google Cloud (Google, 2011-2014)

Motivated by a need to identify future workloads for the Google cloud that are big data and/or big compute, I initiated a Big Science project with the following objectives: (a) drive integration within the Google cloud; (b) identify new value-add capabilities for scientific discovery; (c) make contributions to data-intensive science; and (d) drive revenue. Details in this interview.

Workload characterization of the Google cloud (Google, 2009-11)

The traditional approach to capacity planning and performance tuning in Google is to use traces to drive benchmarks, an approach that provides little ability to do workload forecasting or to exploit workload characteristics in scheduling. I led efforts that: (a) characterized task resource consumption (Sigmetrics Performance Evaluation Review, 2010); (b) demonstrated that task resource consumption is fairly constant during the execution of most long-running, resource intensive tasks (The 5th Workshop on Large Scale Distributed Systems and Middleware, Seattle, 2011); (c) characterized and modeled task placement constraints that restrict which resources tasks can consume (Symposium on Cloud Computing, 2011); and (d) published Google Cluster Data (published ~50 variables for 1 month of data). The results have been used in Google for performance tuning, capacity planning, and scheduler design.

Performance tools for Google's data center operating system (Google, 2010-2011)

Managing Google compute clusters requires tools that can answer what-if questions about the performance impact of changes in hardware, jobs, and scheduling policies. I led a team that built tools used by cluster administrators and service managers to diagnose and resolve problems with scheduling quality. Our contributions include: (a) scaling the tools to be used for all submission of production jobs; (b) developing dashboards that provide actionable explanations to why a job experiences scheduling delays; and (c) inventing and deploying a new technology for what-if analysis of scheduling quality using resource shapes supplied by machines and demanded by jobs.

Performance optimizations of Google's storage infrastructure (Google, 2009-10)

Google's storage stack contains multiple layers, a structure that greatly complicates the identification and resolution of performance problems. Motivated by problems with intermittent long response times (tail latencies) in the storage stack, I formed a team to use the Dapper RPC tracing tool to analyze problems with storage latencies. We developed tools that identified and led to the resolution of performance problems in several key Google storage technologies.

Dynamic estimation of task resource usage in Google's data center operating system (Google, 2009-10)

The Google data center operating system employs a charge-back system to ensure that cloud resources are allocated in accordance with corporate investments. However. jobs often specify resource requirements far in excess of what they actually consume, thereby underutilizing expensive computing resources. Resource estimation provides a mechanism to dynamically adjust resource requirements in a way that makes available resources that can be used speculatively by other jobs. My contributions were to: (a) provide a formal analysis of the resource estimation feedback mechanism (which identified several bugs) and (b) optimize the choice of parameters used in the feedback mechanism to achieve better resource utilizations. Further, I designed and implemented a scheme that provides a two tier system for resource management to provide an explicit mechanism for speculative use of resources.

Management policies for the Google thread manager (Google, 2008)

The requirements for thread management depend on the specifics of the workloads. I/O intensive workloads do better with a high level of concurrency; CPU-intensive workloads prefer lower concurrencies. To address these varied requirements, I modified the Google thread manager to provide a mechanism for user policies for managing thread concurrency levels.

.NET Thread Pool (Microsoft, 2007-8)

The .NET thread pool in version 3.x was a major source of bugs because of the use of complex, interrelated thresholds used to determine the concurrency levels. I developed a thread pool that is based on a simple hill climbing principle that was introduced in .NET 4.0 (Sigmetrics Performance Evaluation Review, 2009). Reports to date are that reliability and performance have improved considerably.

Process optimizations for sustaining engineering in Microsoft's Developer Division (Microsoft, 2006-2007)

This project was motivated by problems at Microsoft with meeting service level expectations for responding to software errors. It turned out that the processes for sustaining engineering in Developer Division involved 7 to 10 different departments with largely informal data exchanges and service expectations. I modeled the process, and identified several optimizations that significantly reduced delays and increased reliability. The results drove the development of a new dashboard for tracking bugs and fixes. Aspects of this work are reported in Network Operations and Management, April, 2008.

Control Engineering of Computing Systems (IBM Research, 2001-2006)

Developed formal methodologies and constructed engineering patterns to enable software practitioners to apply control engineering to computing systems. This work was motivated by the observation that software systems often scale poorly because of deficiencies in the way they handle dynamics, especially changes in workloads and resource characteristics (e.g., failures). The problem of dynamics exists in many engineering disciplines, such as mechanical, electrical, and aeronautic engineering. In these areas, control theory has proved to be an effective tool for analysis and design. The goal of this work has been to achieve similar benefits in engineering software systems.

Prior to this work, there were a number of papers that used control theory to analyze data networks and servers. While these efforts developed interesting techniques, they did not address the objective of providing practical results for professional software engineers. To achieve this objective, three steps were required: (1) identify a simple subset of control theory that can be easily digested by software engineers and provides a good correspondence between model predictions and the empirical observations of dynamics, especially settling times and oscillations; (2) demonstrate that these techniques can provide substantial improvements in commercial products with production workloads; (3) develop educational materials and pedagogy to teach control theory and its application to software practitioners.

The first step, identifying an appropriate subset of control theory, was largely accomplished through a series of "science experiments" on IBM's Lotus Domino Server and the Apache Web Server. We used discrete, deterministic, linear time-invariant (LTI) systems with pole placement design. Our innovations were largely in modeling as described in "Using Control Theory to Achieve Service Level Objectives in Performance Management" (Real Time Systems Journal, 2002, 159 citations) and in system identification for multiple input multiple output (MIMO) models as described in "Using MIMO Feedback Control to Enforce Policies for Interrelated Metrics With Application to the Apache Web Serve" (Network Operations and Management, 2002, 95 citations).

The second step, demonstrating the value of control theory in commercial products, has been an on-going effort for several years. In 2005, IBM shipped the Universal Database Server (DB2) v8.2 that incorporates utilities throttling, a solution to a persistent problem that administrators have with ensuring that utilities such as BACKUP, RESTORE, and REBALANCE make progress but do not cause excessive performance degradation of production work. As detailed in "Throttling Utilities in the IBM DB2 Universal Database Server" (American Control Conference, 2004), our methodology for control engineering played a central role in DB2 v8.2, especially in designing an effective actuator using self-imposed sleep, estimating the impact on production work, and designing a controller that responds quickly but does not oscillate. It turns out that the Microsoft Hotmail team also had challenges with managing administrative work, and used our the DB2 solution for their design.

Release 9.1 of the DB2 product contains another feature in which control engineering plays a central role --- self-tuning memory management. Administrators face considerable challenge with choosing the correct size of buffer pools for mixes of production workloads. In concept, this is a constrained optimization problem in which the objective is to minimize data access times by allocating memory to buffers subject to the constraint of the total size of memory. As discussed in "Incorporating Cost of Control Into the Design of a Load Balancing Controller" (Invited paper, Real-Time and Embedded Technology and Application Systems Symposium, 2004), this problem can be re-cast as a simple regulatory control. This turns out to be a significant advance in control modeling since the approach has broad application to load balancing problems, which are very common in resource management of computing systems.

Last, I have made progress with education on the theory and practice of control engineering for software systems. The starting point was to write a text book for software professionals and researchers--- Feedback Control of Computing Systems (Wiley, 2004, 136 citations). Next, I did tutorials and short classes at ACM Sigmetrics, University of California at Berkeley, and Stanford University. Ultimately, this led to full semester courses at Columbia University (Spring & Fall, 2004) and University of Washington at Seattle (Winter, 2008). 

Key publications are:

  1. Feedback Control of Computing Systems (with Yixin Diao, Sujay Parekh, and Dawn Tilbury), Wiley (2004).
  2. Using MIMO Feedback Control to Enforce Policies for Interrelated Metrics With Application to the Apache Web Serve," Y Diao, N Gandhi, JL Hellerstein, S Parekh, and DM Tilbury. Network Operations and Management, April 15-19 2002, pp. 219-234.
  3. "Using Control Theory to Achieve Service Level Objectives in Performance Management," S Parekh, N Gandhi, JL Hellerstein, D Tilbury, TS Jayram, J Bigus, Real Time Systems Journal, Vol.23, No. 1-2, 2002.
  4. “Achieving Service Rate Objectives With Decay Usage Scheduling,'' IEEE Transactions on Software Engineering, Vol. 19, 1993, 813-825.

Engineering Self-Managing Systems (IBM Research, 2002-2005)

Developed a theory and methodology for quantifying the complexity of system administration thereby enabling a way to assess the value of self-managing systems. Applied this to IBM service offerings and products. Invited Lecturer at the Royal Institute of Technology (Stockholm, Sweden) to initiate their Autonomic Computing Research Center, and keynote speaker at the International Workshop on Self-Organizing Systems. Key publications are:

  1. “A Configuration Complexity Model and Its Application to a Change Management System,” Alexander Keller, Aaron B. Brown, and Joseph L. Hellerstein, IEEE Transactions on Network and Service Management, Vol. 3, No. 1, June, 2007.
  2. "An On-line, Business Oriented Optimization of Performance and Availability for Utility-Based Computing," Joseph L Hellerstien, Kaan Katircioglu, and Maheswaran Surendra. Journal on Selected Areas of Communications, Oct., 2005, Vol. 23, No. 10.
  3. Reducing the Cost of IT Operations---Is Automation Always the Answer? Aaron B. Brown and Joseph L. Hellerstein. HotOS-X, 2005. 

Characterizing Events in Software Services (IBM Research, 1998-2000)

Developed methodologies and tools for detecting patterns of importance to software services and assessing the impact of these patterns on service levels. Validated these approaches by applying them to IBM services and customers. Key publications are:

  1. "A Statistical Approach to Predictive Detection," Joseph L. Hellerstein, Fan Zhang and Perwez Shahabuddin, Computer Networks, January, 2000.
  2. "Mining Partially Periodic Event Patterns With Unknown Periods," S Ma and JL Hellerstein,  International Conference on Data Engineering, 2000.
  3. “Ordering Categorical Data to Improve Visualization," Sheng Ma and Joseph L. Hellerstein, IEEE Symposium on Information Visualization, 1999.


Books

    • Feedback Control of Computing Systems (with Yixin Diao, Sujay Parekh, and Dawn Tilbury), Wiley (2004).
    • Expert Systems in Data Processing (with David A. Klein and Keith R. Milliken), Addison-Wesley (1990).

Other Peer Reviewed Publications

  1. “Some Potential Deadlocks in Layered Communications Architectures'' (with W. W. Chu), Proceedings of the National Computer Conference, 1981, pp. 137-140. 
  2. “The Exclusive Writer Protocol: A Low Cost Approach for Updating Replicated Files in Distributed Real Time Systems'' (with W.W. Chu and M.T. Lan), Proceedings of the Third International Conference on Distributed Computing Systems, 1982, pp. 269-277. 
  3. “Estimation of Intermodule Communication (IMC) and Its Application in Distributed Processing Systems'' (with W.W. Chu and M.T. Lan), IEEE Transactions on Computers, C-33, No. 8, 1984. 
  4. “The Exclusive-writer Approach to Updating Replicated Files in Distributed Processing Systems,'' IEEE Transactions on Computers, C-34, No. 6, 1985, pp. 489-500. 
  5. “YSCOPE: A Shell for Building Expert Systems for Solving Computer-Performance Problems'' (with H. Van Woerkom), Proceedings of the Computer Measurement Group, 1985. 
  6. “YES/MVS and the Automation of Operations for Large Computing Complexes'' (with K.R. Milliken, et al.), IBM Systems Journal, V. 25, No. 2, 1986. 
  7. “YES/L1: Integrating Rule-Based, Procedural, and Real-Time Programming for Industrial Applications'' (with A. Cruise et al.), Proceedings of the IEEE Conference on Artificial Intelligence Applications, February, 1987. 
  8. “An Intuitive Approach to Performance Prediction with Application to Workload Management in VM SP/HPO,'' Proceedings of the Computer Measurement Group, 1987. 
  9. “What's-Different Analysis and its Application to Performance Management in VM SP/HPO,'' Proceedings of the Computer Measurement Group, 1988. 
  10. “VM Performance Analysis: A Worksheet-Based Approach'' (with Robert Berry), Proceedings of the Computer Measurement Group, 1988. (Also appears in the Mainframe Journal, 1989.) 
  11. “A Statistical Approach to Diagnosing Intermittent Performance Problems Using Monotone Relationships,'' Proceedings of ACM Sigmetrics, 1989. 
  12. “Estimating the Utilization of Shared DASD: A Performance Management Perspective'' (with Robert Berry), Proceedings of the Computer Measurement Group, 1989. 
  13. “Obtaining Quantitative Predictions From Monotone Relationships,'' Proceedings of the American Association for Artificial Intelligence, 1990. 
  14. “Fine-Grain Analysis (FGA): A Methodology for Analyzing Intermittent Performance Problems (with Robert Berry, Jeffrey Hedglin, and Paul VanLeer), Proceedings of the Computer Measurement Group, 1990. 
  15. “An Approach to Detecting Changes in the Performance Characteristics of Computer Systems'' (with Robert Berry), Proceedings of ACM Sigmetrics, 1991. 
  16. “Choosing a Service Level Indicator -- Why Not Queue Length?'' (with Robert Berry, Jay Kolb, and Paul VanLeer), Proceedings of the Computer Measurement Group, 1991. 
  17. “Control Considerations for CPU Scheduling in Unix Systems,'' Proceedings of Winter USENIX, 1992. 
  18. “Characterizing and Interpreting Periodic Behavior in Computer Systems'' (with Robert Berry), Proceedings of ACM Sigmetrics, 1992. 
  19. “Obtaining Quantitative Estimates From Monotone Relationships,'' in Recent Advances in Qualitative Physics, Boi Faltings and Peter Struss, Editors, MIT Press, 1992. 
  20. “Controlling CPU Allocations in UNIX Systems,'' Proceedings of the Computer Measurement Group, 1992. 
  21. “A Unified Approach to Interpreting Measurement Data in Performance Management Applications,'' First IEEE Conference on Systems Management, April 14-16, 1993. 
  22. “Modeling Dynamic Behavior With Time Series Analysis,'' Lecture Notes in Computer Science: Performance Evaluation of Computer and Communication Systems, Lorenzo Donatiello and Randolph Nelson (editors), Springer-Verlag, 203-223. 
  23. “A Comparison of Techniques for Diagnosing Performance Problems in Information Systems,'' ACM Sigmetrics, May, 1994, pp. 278-9. 
  24. “How Expert is Your Expert System?" Proceedings of the Computer Measurement Group, 1994. 
  25. “Constructing Quantitative Models Using Monotone Relationships,'' IEEE Transactions on Knowledge and Data Engineering, Vol. 7, No. 2, April, 1995. 
  26. “Automating Performance Management Using Case-Based Reasoning," Computer Measurement Group, 1995. 
  27. “A Flexible and Scalable Approach to Navigating Measurement Data in Performance Management Applications," Robert F. Berry and Joseph L. Hellerstein, Second International Conference on Systems Management, June 19-21, 1996. 
  28. “An Approach To Selecting Metrics for Detecting Performance Problems in Information Systems," Joseph L. Hellerstein, Second International Conference on Systems Management, June 19-21, 1996. 
  29. “Rules of Thumb for Selecting Detection Metrics,'' Proceedings of Computer Measurement Group, December, 1996. 
  30. “Automated Performance Tuning: Possibilities and Realities," Joseph L. Hellerstein, Paper and invited talk at the Computer Measurement Group, Orlando, Florida, December, 1997. 
  31. “Using Multidimensional Databases for Problem Determination and Planning of a Networked Application,'' Third International Conference on Systems Management, April 22-24, 1998. 
  32. “Applications Management--Current Practices, Research Results and Future Directions," Paul Brusil, Joseph Hellerstein, and Hanan Lutfiyya, Journal of Network and Systems Management, Vol. 6, No. 3., 1998. 
  33. “Characterizing Normal Operation of a Web Server: Application to Workload Forecasting and Capacity Planning," Joseph L. Hellerstein, Fan Zhang, and Perwez Shahabuddin, Computer Measurement Group, December, 1998. 
  34. “An Approach to Predictive Detection for Service Management," Joseph L. Hellerstein, Fan Zhang, and Perwez Shahabuddin. Symposium on Integrated Network Management, 1999. 
  35. ``ETE: A Customizable Approach to Measuring End-to-End Response Times and Their Components in Distributed Systems," Joseph L. Hellerstein, Mark Maccabee, W. Nathaniel Mills, and John J. Turek. International Conference on Distributed Computing Systems, 1999. 
  36. “Predictive Models for Proactive Network Management: Application to a Production Web Server," Dongxu Sheng and Joseph L. Hellerstein, Network Operations and Management, 2000. 
  37. “EventBrowser: A Flexible Tool for Scalable Analysis of Event Data," Sheng Ma and Joseph L. Hellerstein, Distributed Operations and Management, 1999. 
  38. “Automated Drill Down: An Approach to Automated Problem Determination for Performance Management," David Hart, Joseph Hellerstein, and Po Yue, Proceedings of the Conference of the Computer Measurement Group, December, 1999. 
  39. “Ordering Categorical Data to Improve Visualization," Sheng Ma and Joseph L. Hellerstein, IEEE Symposium on Information Visualization, 1999. 
  40. "A Statistical Approach to Predictive Detection," Joseph L. Hellerstein, Fan Zhang and Perwez Shahabuddin, Computer Networks, January, 2000. 
  41. "AutoTune: A Generic Agent for Automated Performance Tuning,” JP Bigus, JL Hellerstein, TS Jayram, and MS Squillante, Practical Application of Intelligent Agents and Multi Agent Technology, 2000. 
  42. "Recognizing End-User Transactions in Performance Management," JL Hellerstein, TS Jayram, I Rish, American Association of Artificial Intelligence, 2000. 
  43. "Mining Partially Periodic Event Patterns With Unknown Periods," S Ma and JL Hellerstein, International Conference on Data Engineering, 2000. (Also in Pattern Recognition and String Matching, edited by Dechang Chen and Xiuzhen Cheng, to be published by Kluwer.) 
  44. "Scalable Visualization of Event Data," David Taylor, Nagui Halim, Joseph L Hellerstein, and Sheng Ma, Workshop on Distributed Systems Operations and Management (DSOM), Austin, Texas, December, 2000. 
  45. "An Approach to On-Line Predictive Detection," Fan Zhang and Joseph L. Hellerstein, MASCOTS, 2000. 
  46. "Metrics for Performance Tuning of Web-Based Applications," W. Nathaniel Mills III, LeRoy Krueger, Willy Chiu, Nagui Halim, Joseph L Hellerstein, Mark S Squillante, The Computer Measurement Group, 2000.
  47. "Analysis of Large-Scale Distributed Information Systems", JL Hellerstein, TS Jayram, and MS Squillante, MASCOTS 2000.
  48. "Mining Event Data for Actionable Patterns," JL Hellerstein and S Ma, The Computer Measurement Group, 2000.
  49. "A Systematic Approach to Discovering Correlation Rules for Event Management," L Burns, JL Hellerstein, S Ma, CS Perng, DA Rabenhorst, D Taylor, IFIP/IEEE International Symposium on Integrated Network Management, 2001.
  50. "Event Relationship Networks: A Framework for Action Oriented Analysis in Event Management," D Thoenen, Jim Riosa, JL Hellerstein, RC 21843 and IFIP/IEEE International Symposium on Integrated Network Management, 2001.
  51. "Feedback Control of a Lotus Notes Server: Modeling and Control Design," N. Gandhi, S. Parekh, J. Hellerstein, and D.M. Tilbury, American Control Conference, 2001. (Best paper in session.) 
  52. "EventMiner: An integrated mining tool for scalable analysis of Event data," Sheng Ma, Joseph L. Hellerstein, Chang-sheng Perng, Knowledge and Data Discovery Workshop on Visual Data Mining, 2001. 
  53. "A Business-Oriented Approach to the Design of Feedback Loops for Performance Management," Yixin Diao, Joseph L. Hellerstein, Sujay Parekh, Distributed Operations and Management, 2001. 
  54. "Dependency Analysis in Distributed Systems Using Fault Injection: Application to Problem Determination in an e-Commerce Environment," Saurabh Bagchi, Gautam Kar, and Joseph L. Hellerstein, Distributed Operations and Management, 2001. 
  55. "Rule Induction of Computer Events," Ricardo Vilalta, Sheng Ma, and Joseph L. Hellerstein, Distributed Operations and Management, 2001. 
  56. "FARM: A Framework for Exploring Mining Spaces with Multiple Attributes," Charles Perng, Haixun Wang, Sheng Ma, and Joseph L. Hellerstein, First IEEE Conference on Data Mining, 2001. 
  57. "Mining Mutually Dependent Patterns," Sheng Ma and Joseph L. Hellerstein, IEEE Conference on Data Mining, 2001. 
  58. Managing the Performance of Lotus Notes: A Control Theoretic Approach, N Gandhi, JL Hellerstein, S Parekh, and DM Tilbury, Proceedings of the Computer Measurement Group, 2001. 
  59. Stochastic Modeling of Lotus Notes with a Queueing Model, Yixin Diao, Joseph L. Hellerstein, and Sujay Parekh, Proceedings of the Computer Measurement Group, 2001. 
  60. "Managing Dynamic Services: A Contracts-Based Approach to a Conceptual Architecture," Alexander Keller, Heiko Ludwig, Gautam Kar, Asit Dan, Joseph L Hellerstein. Network Operations and Management, 2002. 
  61. "Mining Mutually Dependent Patterns for System Management," Sheng Ma and Joseph L. Hellerstein, IEEE Journal on Selected Areas in Communications, 2002, pp. 726-735. 
  62. "MIMO Control of an Apache Web Server: Modeling and Controller Design," Y Diao, N Gandhi, JL Hellerstein, S Parekh, and DM Tilbury, American Control Conference, 2002. (Best paper in session.) 
  63. "A General-Purpose Algorithm for Quantitative Diagnosis of Performance Problems," Joseph L. Hellerstein, Journal of Network and Systems Management, June, 2003. 
  64. "Using Fuzzy Control to Maximize Profits in Service Level Management," Y Diao, JL Hellerstein, S Parekh. IBM Systems Journal, Vol 41, No 3, 2002. 
  65. "Case Studies In Prediction of Potential Failures in Computer Systems," R Vilalta, C Apte, JL Hellerstein, S Ma, S Weiss. IBM Systems Journal, Vol 41, No. 3, 2002. 
  66. "Discovering Actionable Patterns in Event Data," JL Hellerstein, S Ma, C Perng. IBM Systems Journal, Vol 41, No 3, 2002. 
  67. User-Directed Exploration of Mining Space With Multiple Attributes", C Perng, H Wang, S Ma, and JL Hellerstein, Knowledge and Data Discovery, 2002. 
  68. "A First-Principles Approach to Constructing Transfer Functions for Admission Control in Computing Systems," JL Hellerstein, Y Diao, and S Parekh. Conference on Decision and Control, 2002. 
  69. "Optimizing Quality of Service Using Fuzzy Control," Y Diao, JL Hellerstein, S Parekh, Distributed Systems Operations and Management, 2002. 
  70. "Managing Web Server Performance with AutoTune Agents," Y Diao, JL Hellerstein, S Parekh, JP Bigus. IBM Systems Journal, Vol 42, No. 1, 2003. 
  71. "Generic On-Line Discovery of Quantitative Models for Service Level Management," Y Diao, F Eskesen, S Froehlich, JL Hellerstein, A Keller, L Spainhower, and M Surendra, IFIP Symposium on Integrated Management, 2003. 
  72. "On-Line Response Time Optimization of An Apache Web Server," Yixin Diao, Xue Lui, Steve Froehlich, Joseph L Hellerstein, Sujay Parekh, and Lui Sha. International Workshop on Quality of Service, 2003. 
  73. "Generic, On-Line Optimization of Multiple Configuration Parameters With Application to a Database Server," Yixin Diao, Frank Eskesen, Steven Froehlich, Joseph L Hellerstein, Lisa Spainhower, and Maheswaran Surendra. IFIP Conference on Distributed Systems Operations and Management, 2003. 
  74. Managing the Performance Impact of Administrative Utilities," Sujay Parekh, Kevin Rose, Joseph L. Hellerstein, Sam Lightstone, Matthew Huras, and Victor Chang. IFIP Conference on Distributed Systems Operations and Management, 2003. 
  75. "Dynamic Surge Protection: An Approach to Handling Unexpected Workload Surges With Resource Actions That Have Lead Times," E. Lassettre, DW Coleman, Y Diao, S Froehlich, JL Hellerstein, L Hsiung, T Mummert, M Raghavachari, G Parker, L Russell, M Surendra, V Tseng, N Wadia, and P Ye. IFIP Conference on Distributed Systems Operations and Management, 2003.. 
  76. "Towards Benchmarking Autonomic Computing Maturity," Sam Lightstone, Joseph Hellerstein, William Tetzlaff, Philippe Janson, Ed Lassettre, Carolyn Norton, Bala Rajaraman, and Lisa Spainhower. IEEE Workshop on Autonomic Computing Principles and Architectures, Banff, Alberta, Canada, 2003. 
  77. "Enforcing Quality of Service Using Decentralized Runtime Feedback Control," Yixin Diao, Bruno Ciciani, Catherine H. Crawford. International Computer Measurement Group Conference, 2003. 
  78. “The CHAMPS System: Change Management with Planning and Scheduling," Alexander Keller, Joseph L. Hellerstein, Joel L. Wolf, Kun-Lung Wu, Vijaya Krishnan. Network Operations and Management, 2004. 
  79. "Using MIMO Linear Control for Load Balancing in Computing Systems," Yixin Diao, Joseph L. Hellerstein, Adam Storm, Maheswaran Surendra, Sam Lightstone, Sujay Parekh, and Christian Garcia-Arellano. American Control Conference, 2004. 
  80. "Challenges in Control Engineering of Computing Systems," Joseph L. Hellerstein, American Control Conference, 2004. 
  81. “Incorporating Cost of Control Into the Design of a Load Balancing Controller,” Yixin Diao, Joseph L. Hellerstein, Adam Storm, Maheswaran Surendra, Sam Lightstone, Sujay Parekh, and Christian Garcia-Arellano. Invited paper, Real-Time and Embedded Technology and Application Systems Symposium, 2004. 
  82. "Throttling Utilities in the IBM DB2 Universal Database Server," Sujay Parekh, Kevin Rose, Yixin Diao, Victor Chang, Joseph L. Hellerstein, Sam Lightstone, Matthew Huras. American Control Conference, 2004. 
  83. “An Approach to Benchmarking Configuration Complexity,” Aaron B. Brown and Joseph L. Hellerstein. SIGOPS 2004. 
  84. “Automating the Provisioning of Application Services with the BPEL4WS Workflow Lanaguage,” Alexander Keller and Remi Badonnel, Distributed Systems Operations and Management (DSOM) 2004. 
  85. “Self-Managing Systems: A Control Theory Foundation,” Yixin Diao, Joseph L. Hellerstein, Gail Kaiser, Sujay Parekh, Dan Phung, OASIS 2004, keynote at High-Speed Local Networks 2004, and IEEE Second conference on Engineering of Autonomic Systems, 2005. 
  86. “Service Level Management: A Dynamic Discovery and Optimization Approach,” Yixin Diao, Frank Eskesen, Steven Froehlich, Joseph L. Hellerstein, Alexander Keller, Lisa F. Spainhower, and Maheswaran Surendra. Electronic Transactions on Network and Systems Management, April, 2005 
  87. "An On-line, Business-Oriented Optimization of Performance and Availability for Utility-Based Computing," Joseph L Hellerstien, Kaan Katircioglu, and Maheswaran Surendra. Journal on Selected Areas of Communications, Oct., 2005, Vol. 23, No. 10. 
  88. A Framework for Applying Inventory Control to Capacity Management for Utility Computing, Joseph L Hellerstein, Kaan Katircioglu, and Maheswaran Surendra. IFIP/IEEE Integrated Management, 2005, pp. 237-250. 
  89. “A Model of Configuration Complexity and Its Application to a Change Management System,” Aaron Brown, Alexander Keller, and Joseph L. Hellerstein. IFIP/IEEE Integrated Management, 2005, pp. 531-644. Best paper in conference. 
  90. “Comparative Studies of Load Balancing With Control and Optimization Techniques,” Yixin Diao, Chai Wah Wu, Joseph L. Hellerstein, Adam J. Storm, Maheswaran Surendra,
  91. Sam Lightstone, Sujay Parekh, Christian Garcia-Arellano, Matthew Carroll, Lee Chu, and Jerome Colaco. American Control Conference, 2005. 
  92. “Control Engineering for Computing Systems,” Joseph L. Hellerstein, Yixin Diao, Sujay Parekh, and Dawn M. Tilbury. IEEE Control Systems Magazine, December 2005, pp. 56-68. 
  93. A Control Theory Foundation for Self-Managing Systems, Yixin Diao, Rean Griffith, Joseph L. Hellerstein, Gail Kaiser, Sujay Parekh, Dan Phung. Journal on Selected Areas of Communications, December, 2005, vol 23, no. 12. 
  94. Control Considerations in Scaling Event Correlation, Wei Xu, Joseph L. Hellerstein, Bill Kramer, and David Patterson. Distributed Systems Operations and Management, 2005. 
  95. “Dynamic Adaptation of Temporal Event Correlation Rules”, Rean Griffith, Joseph L. Hellerstein, Yixin Diao, and Gail Kaiser. To appear in the International Workshop on Quality of Service, 2006. 
  96. “Controlling Quality of Service in Multitier Web Applications”, Yixin Diao, Joseph L. Hellerstein, Sujay Parekh, Hidayatullah Shaikh, Maheswaran Surendra. International Conference on Distributed Computing Systems, 2006. 
  97. “Scaling Control of Computing Systems”, Yixin Diao, Joseph L. Hellerstein, and Sujay Parekh. First International Workshop on Feedback Control Implementation and Design in Computing Systems, 2006. 
  98. “Teaching Control Theory to Computing Practitioners,” Yixin Diao, Joseph L. Hellerstein, and Sujay Parekh. IEEE Advances in Control Education, 2006. 
  99. “Automating System Administration: Landscape, Approaches, and Costs,” Aaron B. Brown, Joseph L. Hellerstein, and Alexander Keller, Handbook of Network and System Administration, edited by Jan Bergstra and Mark Burgess, November/December 2007, ISBN-10: 0-444-52198-4. 
  100. “Control Theory: A Foundational Technique for Self-Managing Databases”, Sam Lightstone, Yixin Diao, Maheswaran Surrendra, Sujay Parekh, Joseph L Hellerstein, Second International Workshop on Self-Managing Database Systems, Instanbul, Turkey, April, 16, 2007. 
  101. “A Configuration Complexity Model and Its Application to a Change Management System,” Alexander Keller, Aaron B. Brown, and Joseph L. Hellerstein, IEEE Transactions on Network and Service Management, Vol. 3, No. 1, June, 2007. 
  102. “Engineering Performance Using Control Theory,” Joseph L. Hellerstein, Yixin Diao, and Sujay Parekh, Computer Measurement Group, December, 2007. 
  103. “Optimizing Software Packages for Application Management,” Joseph L. Hellerstein, Network Operations and Management, April, 2008, Salvador, Bahia, Brazil. 
  104. “Introduction to Control Theory and Its Application to Computing Systems” (with Tarek Abdelzaher, Yixin Diao, Chenyang Lu, and Xiaoyun Zhu), to appear in Lecture Notes for Computer Science, Springer Verlag, 2008. 
  105. “Configuring Resource Managers Using Model Fuzzing: A Case Study of the .NET Thread Pool.” Integrated Management, 2009
  106. “Applying Control Theory in the Real World: Experience With Building a Controller for the .NET Thread Pool,” Joseph L. Hellerstein, Vance Morrison, Eric Eilebrecht. ACM HotMetrics, Seattle, WA June, 2009.
  107. "Control Engineering for Scaling Service Oriented Architectures," Yixin Diao, Joseph L Hellerstein, Sujay Parekh. Accepted book chapter in Methodologies for Non-Functional Requirements in Service Oriented Architecture: Requirements Engineering, Model-Driven Development and Security, Jun Suzuki (ed.), IGI Global.
  108. “Recent Advances in Autonomic Communications,” Raouf Boutaba, J.P. Martin-Flatin, Joseph L. Hellerstein, Randy Howard Katz, George Pavlou. Guest editor editorial for special issue of Journal on Selected Areas of Communications, 2010.
  109. “Research Challenges in Control Engineering of Computing Systems,” JL Hellerstein, S Singhal, and Q Wang. Guest editor editorial for special issue of Journal of Network and Service Management, Vol 6, No. 4, Dec. 2009, pp. 1-6.
  110. “Towards Characterizing Cloud Backend Workloads: Insights from Google Compute Clusters,” Asit Kumar Mishra, Joseph L. Hellerstein, Walfredo Cirne, and Chita Das, ACM Sigmetrics Performance Evaluation Review, 2010.
  111. “Why Feedback Implementations Fail: The Importance of Systematic Testing,” Joseph L. Hellerstein. 5th Workshop on Feedback Control Implementation and Design, Paris, France, 2010.
  112. "Control Engineering for Scaling Service Oriented Architectures," Yixin Diao, Joseph L Hellerstein, Sujay Parekh. Methodologies for Non-Functional Requirements in Service Oriented Architecture: Requirements Engineering, Model-Driven Development and Security, Jun Suzuki (ed.), IGI Global, 2010.
  113. “Modeling and Synthesizing Task Placement Constraints in Google Compute Clusters,” Bikash Sharma, Victor Chudnovsky, Joseph L. Hellerstein, Rasekh Rifaat, and Chita R. Das, ACM Symposium on Cloud Computing, October, 2011.
  114. “Characterizing Task Usage Shapes in Google's Compute Clusters,” Qi Zhang, Raouf Boutaba, and Joseph L. Hellerstein. The 5th Workshop on Large Scale Distributed Systems and Middleware, Seattle, WA, September, 2011.
  115. Science in the Cloud: Accelerating Discovery in the 21st Century,” Joseph L. Hellerstein, Kai Kohlkoff, David Konerding. IEEE Internet Computing, Issue 4, July-Aug, 2012.
  116. "Robust Assessment and Improvement of Breast Cancer Molecular Prognostic Models Through the Sage Bionetworks DREAM Breast Cancer Challenge", Lars Ottestad  Brig Mecham, Ben Sauerwine, Mike Kellen, Lara Mangravite, Matthew Furia, Nicole Deflaux, Bruce Hoff, Xavier Schildwachter, Oscar Rueda, Tyler Pirtle, Joseph Hellerstein,  Samuel Aparicio, Carlos Caldas, Hans Kristian Moen Vollan, Anne-Lise Boerresen-Dale, Science Translational Medicine, 2013.
  117. Dynamic Service Placement in Geographically Distributed Clouds,” Qin Zhang, Raouf Boutaba, Joseph L. Hellerstein, Journal of Selected Areas of Communications, 2013.

  118. Dynamic Heterogeneity-Aware Resource Provisioning in the Cloud”, Qi Zhang , Mohamed Faten Zhani , Raouf Boutaba , Joseph Hellerstein. IEEE Transactions on Cloud Computing, 2014.

Selected Tutorials and Other Presentations

  1. Book presentation:  “Expert Systems in Data Processing”, Computer Measurement Group, 1989.
  2. “An Introduction to Expert Systems for Capacity Management'' (with Robert Berry), Tutorial Presented at ACM Sigmetrics, 1990.
  3. “An Introduction to Expert Systems for Capacity Management'' (with Robert Berry), Proceedings of the Computer Measurement Group, 1990.
  4. “Statistical Modeling for Performance Analysts,'' Tutorial presented at ACM Sigmetrics, 1992.
  5. “Building Regression Models: An Introduction With a Unix Case Study,'' Proceedings of the Computer Measurement Group, December 5-10, 1993, pp. 694-703.
  6. “Management of Complex Information Systems'' (with Seraphin B. Calo and William W. White), NIST-ATP Workshop, Boston, MA, January 31, 1994 and Gaithersburg, MD, August 31, 1994.
  7. “Managing the Virtual Enterprise'' (with Seraphin B. Calo and William W. White), NIST-ATP Workshop, Ann Arbor, Michigan, March 29-30, 1994.
  8. "An Introduction to Control Theory and Its Application to the Design of Information Systems,''  ACM Sigmetrics, May, 1994.
  9. "An Introduction to Statistical Hypothesis Testing,'' Tutorial Presented at ACM, 1995.
  10. “An Introduction to Change-Point Detection,'' Joseph L. Hellerstein, Proceedings of ACM Sigmetrics, June, 1998.
  11. "An Introduction to Control Theory With Applications to Computer Science," JL Hellerstein and S Parekh, ACM Sigmetrics, 2001. 
  12. “Fast Track Introduction to Control Theory for Computer Scientists,” Yixin Diao, Joseph L. Hellerstein, Sujay Parekh, Symposium on Integrated Management, 2003.
  13. Invited Tutorials on Feedback Control of Computing Systems: Integrated Management 2003; Stanford University May, 2003; University of California at Berkeley, 2004; Integrated Management, 2005; ACM Sigmetrics, 2005.
  14. “Lessons Learned from Applying Control Theory to Computing Systems: A Manifesto for Resource Management Engineering,” HP Labs (9/19/2007), IBM Research (10/15/2007), Linkoping University, Sweden.
  15. “Introduction to Control Theory and Its Application to Computing Systems,” ACM Sigmetrics, 2008.
  16. Cloud Computing Panel, NOMS 2008, April 8, 2008, Salvador, Bahia, Brazil.
  17. “QoS is Dead”, Outrageous opinion session, NOMS 2008, April 10, 2008, Salvador, Bahia, Brazil.
  18. “Introduction to Control Theory for Computer Scientists,” (with T Abdelzaher, Y Diao, C Liu, and X Zhao), ACM Sigmetrics Tutorial, June 2, 2008, Annapolis, MD.
  19. “Solving Games of National Importance,” (with Michael Wellman and George Cybenko), ISAT Workshop, August, 2008, Woodshole, MA.
  20. “Optimizing Concurrency Levels in the .NET Thread Pool: A Case Study of Controller Design and Implementation,” Joseph L. Hellerstein, Vance Morrison, Eric Eilebrecht. Feedback Control Implementation and Design in Computing Systems and Networks, Annapolis, MD June, 2008. 
  21. "Recent advances in control theory for network and systems management," International Conference on Integrated Management, Hofstra University, June 2, 2009.
  22. "Making Management Matters Matter," Distinguished Experts Panel, IM 2009, Hofstra University, June 4, 2009.
  23. "Cloud Computing: What Business Models Will Succeed?", IM 2009, Hofstra University, June 3, 2009.
  24. "The Promise of Autonomic Computing -- Has it Delivered?", IM 2009, Hofstra University, June 3, 2009.

Students Advised And Thesis Defenses                                               

  • German S. Goldszmidt, Columbia University. Industry member of thesis committee. Thesis adviser: Yechiam Yemini. November, 1996.

  • Fan Zhang, Columbia University. Industrial Engineering and Operations Research. 2002.

  • Sujay Parekh, University of Washington. 2002-2010Rean Griffith, Columbia University. 2004-2009

  • Mehdi Amirijoo, Linkoping University, Sweden. Invited as thesis opponent. Thesis adviser: Jorgen Hansson. November, 2007.

  • Jin Xiao, University of Waterloo, Computer Science Department. Thesis adviser: Raouf Boutaba, January, 2010.

  • Asit Mishra, Penn State University, Computer Science Department. Thesis advisor: Chita Das. May 2009-2011.
  • Bikash Sharma, Penn State University, Computer Science Department. Thesis advisor: Chita Das. March 2011-present.


Patents

1. “A General Purpose Mechanism for Detecting Performance Problems in Window-Based Systems" ( Robert Berry and Joseph L Hellerstein), issued 3/9/99, 5,881,222. 

2. “A Simple Approach To Case-Based Reasoning for Data Navigation Tasks," (Joseph L Hellerstein) issued 2/10/98, 5,717,835. 

3. “Method and Apparatus for Quantitative Diagnosis of Performance Problems Using External Representations," (Joseph L Hellerstein) 11/30/99, 5,996,090. 

4. “System and Method for Automated Problem Isolation in Systems with Measurements Structured as a Multidimensional Database," (Joseph L. Hellerstein and Po C. Yue), 12/11/2002, 6,330,564. 

5. "Predictive Model-Based Measurement Acquisition," (JL Hellerstein and N Haus), August 6, 2002, US 6,430,615. 

6. " Method and System for Optimal Problem Isolation for Data Structured as a Multidimensional Database" (Joseph L Hellerstein, Tracy Kimbrel, Robert D Kearney, Jayram S Thathachar),

November, 2002, 6,330,564. 

7. “System and methods for using Continuous Optimization for Ordering Categorical Data Sets,” Alina Beygelzimer, Joseph L. Hellerstein, Sheng Ma, and Charles Perng,

September 2, 2003, 6,615,211. 

8. “Methods and Apparatus for Performance Management Using Self-Adjusting Model-Based Policies” issued 1/6/2004, 6,676,128. 

9. “Systems and methods for pairwise analysis of event data” issued 2/24/2004, 6,697,802. 

10. “System and Method for Systematic Construction of Correlation Rules for Event Management,” 6,697,791 on February 24, 2004. 

11. “Method, computer program product, and system for deriving web transaction performance metrics” issued 3/2/2004, 6,701,363. 

12. “Systems and methods for automated navigation between dynamic data with dissimilar structures” issued 3/9/2004, 6,704,721. 

13. “System and method for generic automated tuning for performance management”, issued 4/6/2004, 6,718,358. 

14. “Method, computer program product, and system for deriving web transaction performance metrics”, May, 20004, 6,701,363. 

15. “Object-oriented framework for generic adaptive control,” US 0302611, September 3, 2004. 

16. “Systems and methods for authoring and executing operational policies that use event rates,” US 6792456, September 14, 2004. 

17. “Systems and methods for discovering mutual dependence patterns,” US 6829608, December 7, 2004. 

18. “Systems and methods for exploratory analysis of data for event management,” US 6836894, December 28, 2004. 

19. “Method and system for recognizing end-user transactions,” US6925452, August 2, 2005.“System and method for on-line prediction using dynamic management of

multiple sub-models,” US6937966, August 30, 2005. 

20. “Systems and methods for discovering fully dependent patterns,” US 6990486, January 24, 2006. 

21. “Methods and apparatus for managing computing deployment in the presence of variable workload,” EU 03749287.3-1243-US0327304, February 3, 2006. 

22. “Apparata, articles and methods for discovering partially periodic event patterns,” US 6,996,551, February 7, 2006. 

23. “Systems and methods for service and role-based software distribution,” US 7013461, March 14, 2006. 

24. “Methods and apparatus for performing adaptive and robust prediction,” US 703559, May 2, 2006. 

25. “Adaptive Throttling System for Data Processing Systems,” US 7137019, Nov. 14, 2006.

26.  System and methods for quantitatively evaluating complexity of computing system configuration,” US 7,177,774, February 13, 2007.

27. Method and apparatus for online sample interval determination (with Yixin Diao, and Sam Lightstone), US 8,347,289, January 1, 2013.

Recent Program Committees

  • International Conference on Distributed Computing Systems, 2012.
  • World Wide Web, 2012.
  • IEEE International Conference on Distributed Computing Systems, 2010, Genoa, Italy.
  • Co-editor of IEEE Transactions on Network and Service Management Special Issue on Feedback Control, 2010.
  • Co-editor of Journal of Computer Networks Special Issue on Advances in Systems Management, 2010.
  • Review of “How to Make it Big in Software,” Sam Lightstone, Pearson Press.
  • Workshop on System Management Techniques, Process, and Services (SMTPS), 2010.
  • International Conference on Autonomic Computing, 2010, Washington, DC
  • IEEE/IFIP Distributed Systems Operations and Management, 2009, Venice, Italy.
  • IEEE Communications Magazine, 2009.
  • ACM Hot Metrics, 2009.
  • ICSE-Cloud-09, 2009.
  • ACM GreenMetrics 2009.
  • Co-editor of JSAC Special Issue on Autonomic Communications, 2009.
  • ICC Symposium on Selected Areas of Communications, 2008-2009.
  • IEEE International Journal of Control, May 2008.
  • Distributed Systems Operations and Management, 1996-2008.
  • Integrated Management, 1997, 1999, 2001, 2003, 2005, 2007, 2009.
  • Network Operations and Management, 1998, 2002, 2004, 2006, 2008.
  • Feedback Control Implementation and Design in Computing Systems (FeBID) since 2006-2008.
  • IEEE Automated Network Management (ANM’08) Workshop, 2008.
  • International Workshop on Self-Organizing Systems (IWSOS) 2008, 2009.
  • IEEE Communications Magazine, 2007.
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  

Last modified: December 20, 2012

 
Ċ
Joseph Hellerstein,
Nov 28, 2011, 6:09 AM
Ċ
Joseph Hellerstein,
Nov 28, 2011, 6:25 AM
Ċ
Joseph Hellerstein,
Nov 28, 2011, 6:28 AM
Ċ
Joseph Hellerstein,
Nov 28, 2011, 6:12 AM
Ċ
Joseph Hellerstein,
Nov 28, 2011, 6:18 AM
Ċ
Joseph Hellerstein,
Nov 28, 2011, 5:58 AM
Comments