Robert Hollebeek

Professor
DRL 4N41
(215) 898-9248
(215) 898-2010

 

While studying for his doctorate at UC Berkeley, Robert Hollebeek was a member of one of the first particle physics experiments to use completely digital techniques for acquiring data from fast electronic detectors with hundreds of thousands of electronic elements. This experiment was conducted at the Stanford Linear Accelerator and discovered the charm quark, later leading to the Nobel Prize for Burton Richter.

From the Stanford accelerator, Hollebeek joined Columbia University as a post doctoral researcher and was the lead for computing and electronics from the Columbia group at the CERN Accelerator lab in Geneva Switzerland. From CERN, Hollebeek returned to Stanford as an Assistant Professor, and was involved there in several particle physics experiments with very large scale data and computing needs.

Hollebeek went to the University of Pennsylvania in 1986. The experiment he joined was the largest electronic detector in the world at the time. The research was conducted at the Fermi National Accelerator Laboratory in Chicago, and again involved the collection and analysis of ultra-large data samples. To accomplish this, Hollebeek began designing clustered computing architectures that were in turn used to simulate and analyze physics data. The experiment was one of two detectors responsible for the discovery of the top quark. During this time, Professor Hollebeek became interested in other areas with ultra-large data and computing needs in particular medical images. A pilot program in functional MRI for example used the parallel clustered computing architectures to analyze patterns in brain imaging. At that time, Professor Hollebeek joined an effort to design data storage and index handling for mammography images.

Professor Robert Hollebeek was the technical lead for the National Library of Medicine sponsored National Digital Mammography Archive (NDMA). This digital storage and search/retrieval system funded by the US National Insitutes of Health established a distributed grid of systems for medical records and image storage. Previously, Professor Hollebeek was also co-founder of the National Scalable Cluster Project at the University of Pennsylvania (NSCP); a multi-university consortium that built data mining infrastructure and ultra-large databases connected by high-speed wide area networks. His activities at the University of Pennsylvania included parallel and distributed computing approaches to the analysis of data from many fields including medical data, social and economic data, and scientific data. Common across these areas is the need for ultra-large data systems, data mining, scalable systems, and efficient access to terabyte and petabyte scale data. Dr. Hollebeek and the NSCP group won several national prizes for their work in distributed clustered systems and data mining.

The work on mammography led to the establishment of i3ARCHIVE, Inc. later renamed NDMA, which provides distributed grid systems for medical storage and records handling. Hollebeek was the Chief Technology Officer for NDMA and expanded the functionality of the original systems into other areas of medical records handling, new grid systems for electronic health records access, established physical and network infrastructure to support NDMA data systems, added new retrieval and search mechanisms, and helped create new products in the area of disaster recovery and data mining in medical records.

 

Education: 

PhD, University of California, Berkeley (1974)

A. B., B.S. in Physics, Calvin College (1970)

Research Interests: 

Data-mining applied to "big" science

Selected Publications: 

 

  • NSF Workshop on New Visions for Large Scale Networks: Research and Applications. R. Hollebeek, M. Schnall, B. Beckerman, "Petabyte Scale Radiology Archiving and Retrieval", Mar, 2001.
  • “The National Scalable Cluster Project: Three lessons about high performance data mining and data intensive computing”, R. Hollebeek and R. Grossman, HANDBOOK OF MASSIVE DATA SETS, Kluwer Academic Publishers
  • Five Most Important Physics Publications (of over four hundred) Observation of top quark production in anti-p p collisions.
    By CDF Collaboration (F. Abe et al.). Mar 1995.
    Published in Phys.Rev.Lett.74:2626-2631,1995.
  • Heavy Quarks, Experimental.
    By Robert Hollebeek (Penn U.), UPR-0188E, Feb 1990. 66pp.
    Published in SLAC Summer Inst.1989:1-44 (QCD161:S76:1989)
  • Search for Anomalous Single Photon Production at PEP.
    By G. Bartha, et al., SLAC-PUB-3817, Oct 1985. 14pp.
    Published in Phys.Rev.Lett.56:685,1986.
  • Disruption Limits for Linear Colliders.
    By R.J. Hollebeek (SLAC), SLAC-PUB-2535, Jun 1980. 41pp.
    Published in Nucl.Instrum.Meth.184:333,1981.
  • Discovery of a Narrow Resonance in e+e- annihilation.
    By J.E. Augustin, et al. SLAC-PUB-1504, Nov 1974. 6pp. 

 

Courses Taught: 

Phys 140/150: Principles I (Intro Physics)