CV

Ryan Marcus

Email: ryan@cs.brandeis.edu
Website: http://ryanmarc.us

Research interests

Cloud computing, distributed database systems, high performance computing, programming languages.

Education

Ph.D. Candidate, Brandeis University 2015-now. Computer science.
- Expected graduation date: May 2019
- Advisor: Olga Papaemmanouil
M.A. Brandeis University, 2014-2015. Computer science, 3.9 GPA
B.S. University of Arizona Honors College, 2010-2014. Cum laude and honors. Computer science, mathematics, gender and women’s studies.

Experience

Ph.D. Candidate, Brandeis University 2014 - now.
- Researching cloud database systems, focus on applying machine learning algorithms to query scheduling as well as other classically hard problems (with Dr. Olga Papaemmanouil)
Software developer, HPE Vertica 2015.
- Worked on low-level query execution engine. Achieved a 4x performance boost for rollup analytic queries
- Added counters and assertions to fix data corruption bugs
Los Alamos National Laboratory 2009-2015.
- Developed a machine-learning framework for automatic performance analysis (with Cornell Wright)
- Parallelized serial algorithms on GPUs. Designed novel median filter approximation algorithm, improved 3D image reconstruction performance by 3x (with Dr. William Ward)
- Implemented exascale co-design neutron transport code, presented at SC10+SC11 (with Dr. Larry Cox)

Publications

Referred publications

Ryan Marcus and Olga Papaemmanouil. “Releasing Cloud Databases from the Chains of Performance Prediction Models.” Conference on Innovative Data Systems Research (CIDR) 2017 (to appear).

Ryan Marcus and Olga Papaemmanouil. “WiSeDB: a learning-based workload management advisor for cloud databases.” Proceedings of the VLDB Endowment, Volume 9, Issue 10, June 2016. (pdf)

Ryan Marcus and Olga Papaemmanouil. “Workload Management for Cloud Databases via Machine Learning.” Workshop on Cloud Data Management and the IEEE International Conference on Data Engineering, CloudDM 2016. (pdf)

Other publications

Ryan Marcus. “Techniques for Automated Performance Analysis.” Tech. Rep. LA-UR-14-26577, Los Alamos National Laboratory, 2014. (pdf)

Ryan Marcus and William Ward. “DP: A fast median filter approximation.” Tech. Rep. LA-UR-13-25331, Los Alamos National Laboratory, 2013. (pdf)

Ryan Marcus. “MCMini: Monte Carlo on GPGPU.” Tech. Rep. LA-UR-12-23206, Los Alamos National Laboratory, 2012. (pdf)

Lawrence Cox and Ryan Marcus. “Developing a Monte Carlo mini-app for exascale co-design.” Tech. Rep. LA-UR-11-06085, Los Alamos National Laboratory, 2011. (pdf)

Presentations

Ryan Marcus. “Machine Learning for Cloud Database Workload Management,” invited to Hewlett Packard Enterprise (HPE) Vertica ; 2016.

Ryan Marcus. “Machine Learning for Humans,” invited for the Los Alamos National Laboratory Data Science Summer School ; 2015.

Ryan Marcus. “Shared Memory for Many-Core: A Hydrodynamics Case Study” for the Los Alamos National Laboratory Research Symposium, LA-UR-15-25303 ; 2015.

CV