Recent News

I successfully defended my PhD in computer science on November 22, 2016. I'd like to thank my adviser Dean Tullsen, my committee, and the countless others that helped me throughout the journey.

My paper on cache and memory bandwidth conserving, high-performance hash tables was accepted at USENIX ATC'16. The paper presents the Horton table, a high-performance hash table that is designed to more efficiently utilize hardware cache and memory bandwidth by reducing the mean number of cache lines that need to be accessed when performing a key lookup on the table. It does so without sacrificing the SIMD friendliness (necessary for peak performance and energy effciency), table occupancy (a.k.a. table fullness or load factor), or worst-case lookup cost in cache lines of the prior state-of-the-art (a bucketized cuckoo hash table that employs 2-choice cuckoo hashing). The paper, slides, and presentation audio are openly accessible on the conference website at this link.

I am working on a DOE FF2 project at AMD Research. We are exploring architectural and software tradeoffs for a future processor in memory architecture.

Publications

For my citation counts, metrics and those of my co-authors, check out my Google Scholar profile.

Alex D. Breslow, Dong Ping Zhang, Joseph L. Greathouse, Nuwan Jayasena, and Dean M. Tullsen. Horton Tables: Fast Hash Tables for In-Memory Data-Intensive Computing. In the Proceedings of USENIX ATC'16. [pdf | slides | bib | ATC'16 talk audio]

Alex D. Breslow, Ananta Tiwari, Martin Schulz, Laura Carrington, Lingjia Tang, and Jason Mars. Enabling Fair Pricing on HPC Systems with Node Sharing. In the Proceedings of SC'13. Best Paper Finalist and Best Student Paper Finalist

[pdf | bib]

Hailong Yang, Alex Breslow, Jason Mars, and Lingjia Tang. Bubble-Flux: Precise Online QoS Management for Increased Utilization in Warehouse Scale Computers. In the Proceedings of ISCA'13. [pdf | bib]

Alex D. Breslow, Leo Porter, Ananta Tiwari, Michael A. Laurenzano, Laura Carrington, Dean M. Tullsen, and Allan E. Snavely. The Case for Colocation of HPC Workloads. Concurrency and Computation: Practice and Experience; Special issue on the Analysis of Performance and Power for Highly Parallel Systems (CCPE 2013). [pdf | bib]

Andrew Danner, Jake Baskin, Alexander Breslow, and David Wilikofsky. Hybrid MPI/GPU Interpolation for Grid DEM Construction. In Proc. ACM Symposium on Advances in Geographic Information Systems, pages 299-308, 2012. [pdf | bib]

Research Statement

I primarily work in the areas of parallel and distributed computing, computer architecture, and computer systems. Most of my work has focused on improving the performance, energy efficiency, and utilization of data centers and supercomputers. Recently, I have been doing work on bottleneck analysis of core primitives from columnar in-memory databases on GPUs as well as optimizing hash tables for bandwidth-constrained systems. This more recent work has been conducted at AMD Research as part of a U.S. Department of Energy Fast Forward2 project on near-data computing.

My advisor is Dean Tullsen.

Side Interests

Aside from research, I am an avid runner and athlete. I also enjoy playing tennis, rock climbing, lifting weights and practicing yoga. Outside of fitness, I speak Spanish and French with relative fluency and enjoy employing these language skills when traveling in Europe and Latin America.

Acknowledgements

Special thanks to my mentors and collaborators throughout the years