Scott A. Brandt

Dr. Scott A. Brandt is Professor of Computer Science at the University of California, Santa Cruz. He is also Director of the UCSC Systems Research Laboratory, co-founder of the UCSC Storage Systems Research Center and, co-founder and Director of the UCSC/Los Alamos Institute for Scalable Scientific Data Management. Dr. Brandt's research is in the area of Computer Systems. He specializes in Storage Systems and Real-Time Systems and, more recently, System and Storage Performance Management. His Storage System research includes high-performance peta-scale object-based storage and the use of new storage technologies to improve storage system performance and reliability. His real-time research focuses on integrating real-time and non-real-time processing into a uniform processing environment. His performance management research integrates the two to provide processing and I/O performance guarantees in local and distributed systems.

Dr. Brandt joined UC Santa Cruz in 1999. Prior to that he spent a number of years doing research and development in industry, including secure operating systems research at Secure Computing Corporation, and real-time image processing systems research at Alliant Techsystems Research and Technology Center and Honeywell Systems and Research Center. Dr. Brandt also co-founded Theseus Research, a small company devoted to researching asynchronous circuit technology and parallel computer languages. Dr. Brandt received his Ph.D. in Computer Science from the University of Colorado at Boulder in 1999, his M.S. in Computer Science from the University of Minnesota in 1994, and his Bachelor of Mathematics from the University of Minnesota in 1987.

Research

Dr. Brandt's research is in the broad area of Computer Systems. He specializes in Storage Systems, Real-Time Systems and Performance Management. He also does research in other systems-related areas. A comprehensive (but somewhat out of date) list of his publications can be found here.

Recent and Notable

• Ceph is in Linux!
  See the press release and the official Ceph website for more information.

• Roberto Pineiro, Kleoni Iouannidou, Carlos Maltzahn, and Scott Brandt, "RAD-FLOWS: Buffering for Predictable Communication," The IEEE Real-Time and Embedded Technology and Applications Symposium, Chicago, Illinois, April 11–14, 2011.

• Greg Levin, Shelby Funk, Caitlin Sadowski, Ian Pye, and Scott A. Brandt, ”DP-FAIR: A Simple Model for Understanding Optimal Multiprocessor Scheduling," The Euromicro Conference on Real-Time Systems (ECRTS), Brussels, Belgium, July–9, 2010. Best Paper Award

• Anna Povzner, Darren Sawyer, and Scott A. Brandt, ”Horizon: Efficient Deadline-Driver Disk I/O Management for Distributed Storage Systems,"” The ACM International Symposium on High Performance Distributed Computing, Chicago, Illinois, June 20–25,
  2010. Best Paper Award.

• Changkyu Kim, Jatin Dhhugani, Nadathur Satish, Eric Sedlar, Anthony Nguyen, Tim Kaldewey, Victor Lee, Scott Brandt, and Pradeep Dubey,”FAST: Fast Architecture Sensitive Tree Search on Modern CPUs and GPUs” ACM SIGMOD, Indianapolis, Indiana, June 6–11,
  2010. Best Paper Award.

• Tim Kaldewey, Anna Povzner, Theodore Wong, Richard Golding, Scott Brandt, Carlos Maltzahn, "Virtualizing Disk Performance", The IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS 2008), St. Louis, Missouri, April 22–24, 2008. Springer Journal of Real-Time Systems Award for Best Student Paper.

• David Bigelow, Scott A. Brandt, John Bent, H. B. Chen, "Mahanaxar: Quality of Service Guarantees in High-Bandwidth, Real-Time Streaming Data Storage," The IEEE / NASA Conference on Mass Storage Systems and Technologies (MSST 2010), Lake Tahoe, Nevada, May 3–7, 2010.

• Timothy Bisson, Scott A. Brandt, Darrell Long, "A Hybrid Disk-Aware Spin-Down Algorithm with I/O Subsystem Support," The IEEE International Performance, Computing, and Communications Conference (IPCCC 2007), New Orleans, Louisiana, April 11–13, 2007. Best Paper Award.

• Sage Weil, Scott A. Brandt, Ethan Miller, Darrell Long, and Carlos Maltzahn, ``Ceph: A Scalable, High-Performance, Distributed Object-based Storage System,'' Symposium on Operating Systems Design and Implementation (OSDI '06), pp. 307--320, Seattle, Washington, November 6–8. 2006.

Storage Systems

I/O bandwidth is increasing arithmetically while CPU performance, disk capacity, and I/O requirements are all increasing exponentially. The ever-widening performance gap this creates both demands and enables increasingly intelligent data storage and retrieval. Our research examines high-performance data storage storage architectures, algorithms, and technologies aimed at addressing this performance gap.

Projects:

• High-performance peta-scale object-based storage
• MRAM-based storage
• MEMS-based storage
• Adaptive Caching
• Prediction and Grouping

Representative publications:

• Sage A. Weil, Scott A. Brandt, Carlos Maltzahn, and Ethan L. Miller, “CRUSH: Controlled, Scalable, and Decentralized Placement of Replicated Data,” International Conference for High Performance Computing, Networking, Storage, and Analysis (SC06), Denver, Colorado, November 11–17, 2006.

• Sage A. Weil, Scott A. Brandt, Ethan L. Miller, Darrell D. E. Long, and Carlos Maltzahn, “Ceph: A Scalable, High-Performance, Distributed Object-based Storage System,” Symposium on Operating Systems Design and Implementation (OSDI ’06), Seattle, Washington, November 6–8. 2006.

• Bo Hong, Scott A. Brandt, Darrell D. E. Long, Ethan L. Miller, and Ying Lin, “Using MEMS-Based Storage in Computer Systems–Device Modeling and Management,” ACM Transactions on Storage, Volume 2, No. 2, pp. 1–22, May 2006.

• Bo Hong, Feng Wang, Scott A. Brandt, Darrell D. E. Long, and Thomas J. E. Schwarz, “Using MEMS-Based Storage in Computer Systems–MEMS Storage Architectures,” ACM Transactions on Storage, Volume 2, No. 1, pp. 1–21, February 2006.

• Sage A. Weil, Kristal T. Pollack, Scott A. Brandt, and Ethan L. Miller, "Dynamica Metadata Management for Petabyte-scale File Systems," International Conference for High Performance Computing, Networking, Storage, and Analysis (SC2004), November, 2004.

Real-Time Systems

We are investigating a new model for dynamic and fully integrated real-time scheduling called Resource Allocation/Dispatching(RAD). RAD allows for the detailed management of both the resources consumed and the timing of the delivery of those resource for each executing process. RAD fully characterizes the timeliness requirements of the processes, allowing each process to have different requirements with respect to both of these quantities and enabling resource delivery precisely tailored to the needs of each process. The RAD model of CPU resource management has led to the development of several distinct CPU resource managers: RBED, HodgePodge, BEST, BeRate, and DQM. We have also applied this model to control system and, more recently, to storage Quality of Service.

Projects:

• Integrated real-time and non-real-time processing
• Storage Quality of Service
• DQM

Representative publications:

• Caixue Lin, Tim Kaldewey, Anna Povzner, Scott A. Brandt, ”Diverse Soft Real-Time Processing in an
  Integrated System,” IEEE Real-Time Systems Symposium (RTSS 2006), Rio de Janeiro, Brazil, December 5–8,
  2006.

• Caixue Lin and Scott A. Brandt, "Improving Soft Real-Time Performance Through Better Slack Management," IEEE Real-Time Systems Symposium (RTSS 2005), pp. 3–14, December, 5–8, 2005.

• Joel Wu, Scott Banachowski, and Scott A. Brandt, "Hierarchical Disk Sharing for Multimedia Systems," ACM International Workshop on Network and Operating System Support for Digital Audio and Video (NOSSDAV 2005), pp. 189–194, June 2005.

• Scott Banachowski, Timothy Bisson, and Scott A. Brandt, "Integrating Best-effort Scheduling into a Real-Time System," IEEE Real-Time Systems Symposium (RTSS 2004), December, 2004.

• Pau Marti, Caixue Lin, Scott Brandt, Manel Velasco, and Josep Fuertes, "Optimal State Feedback Resource Allocation for Resource-Constrained Control Tasks," IEEE Real-Time Systems Symposium (RTSS 2004), pp. 161–172, December, 2004.

Teaching

Dr. Brandt teaches both undergraduate and graduate courses. His undergraduate course offerings include CMPS 12A: Introduction to Programming, CMPS 12B: Data Structures, CMPS 13H: Honors Introduction to Programming and Data Structures, CMPS 105: Systems Programming, and CMPS 111: Introduction to Operating Systems. His graduate course offerings include CMPS 221: Advanced Operating Systems, and CMPS 290S: Advanced Topics in Computer Systems (covering Storage Systems and Real-Time Systems in different offerings).

Class Web Pages:

Current:

• CMPS 221—Advanced Operating Systems

All:

• CMPS 12A—Introduction to Programming

• CMPS 13H—Honors Introduction to Programming and Data Structures

• CMPS 111—Introduction to Operating Systems

• CMPS 200 - Introduction to Research and Teaching in Computer Science and Engineering and Electrical Engineering and Applied Math and Statistics and Technology and Information Management
  

• CMPS 221—Advanced Operating Systems

• CMPS 280S: Seminar in Computer Systems

• CMPS 290S—Advanced Topics in Computer Systems: Storage Systems

• CMPS 290S—Advanced Topics in Computer Systems: Real-Time and Soft Real-Time Systems

Current and Former Graduate Students: