Project Description:

Recent evaluation studies on modern processors show that database workloads do not take advantage of the sophisticated processor and memory architectures as much as scientific workloads do. This is counter-intuitive, because many modern database applications (e.g., data mining) are compute and memory intensive. I have started investigating this problem by examining the behavior of four commercial database management systems running microbenchmarks and TPC on an Intel platform to find common performance bottlenecks. The striking common result is that, even when running microbenchmarks, DBMSs spend most of their execution time on memory stalls. In addition, the dominant memory bottlenecks are due to data misses on the second-level cache and misses on the first-level instruction cache. My research focuses on techniques to alleviate performance bottlenecks throughout the memory hierarchy (including caches, memory, and disks).

People:

Faculty:

• Anastasia Ailamaki

Graduate Students:

• Mengzhi Wang
• Shimin Chen
• Minglong Shao
• Nikolaos Hardavellas
• Christopher Colohan
• Ippokratis Pandis

Undergraduate Students:

• Jose Morales

Publications: (by year)

2008

• Critical Sections: Re-emerging Scalability Concerns for Database Storage Engines
  R. Johnson, I. Pandis, and A. Ailamaki.
  In proceedings of the 4th DaMoN, Vancouver, Canada, June 2008.
  PDF


• Shore-MT: A Quest for Scalability in the Many-Core Era
  R. Johnson, I. Pandis, N. Hardavellas, and A. Ailamaki.
  Carnegie Mellon University Technical Report CMU-CS-08-114, April 2008.
  


• Incrementally Parallelizing Database Transactions with Thread-Level Speculation
  C.B. Colohan, A. Ailamaki, J.G. Steffan, and T.C. Mowry.
  ACM Transactions on Computer Systems (TOCS), February 2008.
  PDF

2007

• MultiMap: Preserving Disk Locality for Multidimensional Datasets
  M. Shao, S. Papadomanolakis, S. Schlosser, J. Schindler, A. Ailamaki, and G.R. Ganger.
  In Proceedings of the International Conference of Data Engineering (ICDE), 2007.
  PDF


• Database Servers on Chip Multiprocessors: Limitations and Opportunities
  N. Hardavellas, I. Pandis, R. Johnson, N. Mancheril, A. Ailamaki, and B. Falsafi.
  In Proceedings of the 3rd CIDR, Asilomar, CA, January 2007.
  PDF


• CMP Support for Large and Dependent Speculative Threads
  C.B. Colohan, A. Ailamaki, J.G. Steffan, and T.C. Mowry.
  IEEE Transactions on Parallel and Distributed Systems (TPDS), Volume 18 Issue 8, August 2007.
  PDF


• Improving Hash Join Performance through Prefetching
  S. Chen, A. Ailamaki, P.B. Gibbons, and T.C. Mowry.
  ACM Transactions on Database Systems (TODS), September 2007.
  PDF


• To Share Or Not To Share?
  R. Johnson, N. Hardavellas, I. Pandis, N. Mancheril, S. Harizopoulos, K. Sabirli, A. Ailamaki, and B. Falsafi.
  In Proceedings of the 33th VLDB, Vienna, Austria, September 2007.
  PDF


• Mechanisms for Store-wait-free Multiprocessors
  T.F. Wenisch, A. Ailamaki, B. Falsafi, and A. Moshovos..
  In Proceedings of the 34th International Symposium on Computer Architecture (ISCA), June 2007.
  PDF

2006

• Tolerating Dependences Between Large Speculative Threads Via Sub-Threads
  C.B. Colohan, A. Ailamaki, J.G. Steffan, and T.C. Mowry.
  In Proceedings of the International Symposium on Computer Architecture (ISCA2006). Boston, MA, June 2006.
  PDF


• Parallel Depth First vs. Work Stealing Schedulers on CMP Architectures
  V. Liaskovitis, S. Chen, P.B. Gibbons, A. Ailamaki, G.E. Blelloch, B. Falsafi, L. Fix, M. Kozuch, T.C. Mowry, and C. Wilkerson.
  In Proceedings of the 18th ACM Symposium on Parallelism in Algorithms and Architectures (SPAA06).
  PDF

2005

• Store-Ordered Streaming of Shared Memory
  T.F. Wenisch, S. Somogyi, N. Hardavellas, J. Kim, C. Gniady, A. Ailamaki, and B. Falsafi.
  In Proceedings of the 14th International Conference on Parallel Architectures and Compilation Techniques (PACT), September 2005.
  PDF


• On Multidimensional Data and Modern Disks
  S. Schlosser, J. Schindler, S. Papadomanolakis, M. Shao, A. Ailamaki, C. Faloutsos, and G.R. Ganger
  In Proceedings of the Fourth USENIX Conference on File and Storage Technologies (FAST), 2005.
  PDF


• DBMbench: Fast and Accurate Database Workload Representation on Modern Microarchitecture
  M. Shao, A. Ailamaki, and B. Falsafi.
  Conference of the Centre for Advanced Studies on Collaborative Research 2005.
  Abstract / PDF


• Optimistic Intra-Transaction Parallelism on Chip Multiprocessors
  C.B. Colohan, A. Ailamaki, J.G. Steffan, and T.C. Mowry.
  In Proceedings of the 31st International Conference on Very Large Data Bases (VLDB 2005), Trondheim, Norway, Aug. 2005.
  PDF


• Temporal Streaming of Shared Memory
  T.F. Wenisch, S. Somogyi, N. Hardavellas, J. Kim, A. Ailamaki, and B. Falsafi.
  In Proceedings of the 32nd Annual ACM/IEEE International Symposium on Computer Architecture (ISCA), June 2005.
  PDF

2004

• Storage Device Performance Prediction with CART Models
  M. Wang, K. Au, A. Ailamaki, A. Brockwell, C. Faloutsos, and G.R. Ganger.
  In Proceedings of the International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS), Volendam, The Netherlands, October 2004.
  Abstract / PDF


• Clotho: Decoupling Memory Page Layout from Storage Organization
  M. Shao, J. Schindler, S.W. Schlosser, A. Ailamaki, and G.R. Ganger.
  In Proceedings of the 30th International Conference on Very Large Data Bases, Toronto, Canada, August 2004.
  Abstract / PDF


• Memory Coherence Activity Prediction in Commercial Workloads
  S. Somogyi, T.F. Wenisch, N. Hardavellas, J. Kim, A. Ailamaki, and B. Falsafi.
  In Proceedings of the Workshop on Memory Performance Issues (WMPI), Munich, Germany, June 2004.
  Abstract / PDF


• Atropos: A Disk Array Volume Manager for Orchestrated Use of Disks
  J. Schindler, S.W. Schlosser, M. Shao, A. Ailamaki, and G.R. Ganger.
  In Proceedings of the Third USENIX Conference on File and Storage Technologies (FAST), San Francisco, CA, March 2004.
  Abstract / PDF


• Diamond: A Storage Architecture for Early Discard in Interactive Search
  L. Huston, R. Sukthankar, R. Wickremesinghe, M. Satyanarayanan, G.R. Ganger, E. Riedel, and A. Ailamaki.
  In Proceedings of the Third USENIX Conference on File and Storage Technologies (FAST), San Francisco, CA, March 2004.
  Abstract / PDF

2003

• Exposing Undergraduate Students to Database System Internals
  A. Ailamaki and, J.M. Hellerstein.
  Sigmod Record 32(3), September 2003.
  Abstract / PDF


• Lachesis: Robust Database Storage Management Based on Device-specific Performance Characteristics
  J. Schindler, A. Ailamaki, and G. R. Ganger.
  In Proceedings of the 29th International Conference on Very Large Data Bases (VLDB), Berlin, Germany, September 2003.
  PDF


• DBMbench: Microbenchmarking Database Systems in a Small, yet Real World
  M. Shao, and A. Ailamaki.
  Carnegie Mellon University Technical Report CMU-CS-03-161, 2003.
  Abstract / PDF

2002

• Data Page Layouts for Relational Databases on Deep Memory Hierarchies
  A. Ailamaki, D. J. DeWitt, and M. D. Hill.
  The VLDB Journal 11(3), 2002.
  Abstract / PDF


• Capturing the Spatio-Temporal Behavior of Real Traffic Data
  M. Wang, A. Ailamaki, and C. Faloutsos.
  Best student paper award in Proceedings of the 22nd edition of the IFIP WG 7.3 International Symposium on Computer Modeling, Measurement and Evaluation (Performance), Roma, Italy, September 2002.
  Abstract / PDF

2001

• Weaving Relations for Cache Performance
  A. Ailamaki, D.J. DeWitt, M.D. Hill, and M. Skounakis.
  Best paper award in Proceedings of the 27th International Conference on Very Large Data Bases (VLDB), Roma, Italy, September 2001.
  PDF


• Walking Four Machines By The Shore
  A. Ailamaki, D.J. DeWitt, and M.D. Hill.
  Fourth Workshop on Computer Architecture Evaluation using Commercial Workloads (CAECW'01), January 2001.
  

Presentations:

Please contact the PI for copies of the slides.

• "Multicore: Friend or Foe? Multicore Architecture, Scalability"
• "To Share or Not To Share?"
• "Database Servers on Chip Multiprocessors: Limitations and Opportunities"
• "Weaving Relations for Cache Performance"