Dr. Edgar A. Leon
Computer Scientist at
Lawrence
Livermore National Laboratory
7000 East Ave L-561, Livermore, CA 94550
My
ORCID
Research interests
Operating systems, distributed systems, networks, and computer
architecture. In particular, identifying and addressing software
challenges posed by emerging architectures in regards to
performance, power, and usability. Research areas include
performance analysis of parallel applications; high-performance
communication interfaces and protocols for system area networks;
and simulation of cluster systems.
Biosketch
At LLNL I conduct research on power-aware computing, emerging
memory technologies, system noise, and resilience for exascale
machines. I was a member of the Novel Systems Architecture group
at IBM Research working on performance analysis and optimization
for PERCS systems. Before IBM, I worked for Sandia
National Laboratories where I developed a scalable simulation
environment to study the impact of novel architectures on the
performance of MPI applications. I received my M.S. and
Ph.D. degrees in computer science from
the University of New
Mexico (UNM) under the supervision of Dr. Arthur
B. Maccabe. In my dissertation, I evaluated the impact of cache
injection of incoming network messages on parallel application
performance and collective operations to address the memory wall
for I/O. For my master's thesis, I created a tool to measure
parallel application sensitivity to variation in communication
parameters based on the LogGP model of computation. During my
graduate work, I interned at IBM T. J. Watson Research
Center and
Intel Santa Clara Engineering Computing.
Selected Publications
- ISC 2019
- Edgar A. Leon.
Providing Cross-Architecture Affinity of Parallel
Applications on Emerging Systems.
International Supercomputing Conference;
Research Poster, June 2019.
- PASC 2019
- Edgar A. Leon (organizer).
Mapping Parallel Scientific Applications onto Complex
Architectures Portably and Efficiently.
ACM Platform for Advanced Scientific Computing conference;
Minisymposium, June 2019.
- MEMSYS 2018
- Edgar A. Leon and Matthieu Hautreux.
Achieving Transparency Mapping Parallel Applications: A
Memory Hierarchy Affair.
ACM International Symposium on Memory Systems,
October 2018.
- INFOCOMP 2018
- Bo Li, Edgar A. Leon, and Kirk W. Cameron.
Understanding Power Measurement Capabilities on Zaius
Power9.
IARIA International Conference on Advanced Communications and
Computation,
July 2018. Best Paper Award.
- GTC 2018
- Edgar A. Leon.
Mapping MPI+X Applications to Multi-GPU Architectures: A
Performance-Portable Approach.
GPU Technology Conference,
March 2018.
- SC 2017
- Nikhil Jain, Abhinav Bhatele, Louis Howell, David Boehme,
Ian Karlin, Edgar A. Leon, Misbah Mubarak, Noah Wolfe, Todd
Gamblin, and Matthew Leininger.
Predicting the Performance Impact of Different Fat-Tree
Configurations.
IEEE/ACM International Conference for High Performance
Computing, Networking, Storage and Analysis,
November 2017.
- MEMSYS 2017
- Edgar A. Leon.
mpibind: A Memory-Centric Affinity Algorithm for
Hybrid Applications.
ACM International Symposium on Memory Systems,
October 2017.
- HPDC 2017
- Bo Li, Edgar A. Leon, and Kirk W. Cameron.
COS: A Parallel Performance Model for Dynamic Variations in
Processor Speed, Memory Speed, and Thread Concurrency.
ACM International Symposium on High-Performance Parallel and
Distributed Computing,
June 2017. Karsten Schwan Best Paper Award.
- INFOCOMP 2017
- Edgar A. Leon, Chris Chambreau, and Matthew
L. Leininger.
What Do Scientific Applications Need? An
Empirical Study of Multirail Network Bandwidth.
IARIA International Conference on Advanced Communications and
Computation,
June 2017. Best Paper Award.
- SC 2016
- Edgar A. Leon, Ian Karlin, Abhinav Bhatele, Steven
H. Langer, Chris Chambreau, Louis H. Howell, Trent D'Hooge, and
Matthew L. Leininger.
Characterizing Parallel Scientific Applications on
Commodity Clusters: An Empirical Study of a Tapered
Fat-Tree.
IEEE/ACM International Conference for High
Performance Computing, Networking, Storage and Analysis,
November 2016.
- SC 2016
- Ryan J. Huber and Edgar A. Leon.
Evaluating Best and Worst Case Scenarios on Two-Level
Memory Systems.
IEEE/ACM International Conference for High
Performance Computing, Networking, Storage and Analysis;
Research Poster, November 2016.
- PARCO 2016
- Edgar A. Leon, Ian Karlin, Ryan E. Grant, and Matthew
Dosanjh.
Program optimizations: The interplay between power,
performance, and energy.
Parallel Computing journal, October 2016.
- ISC 2016
- Edgar A. Leon.
Infrastructure for Evaluating Emerging Multi-level,
Heterogeneous Memory Systems.
International Supercomputing Conference; Research Poster, June
2016.
- IPDPS 2016
- Edgar A. Leon, Ian Karlin, and Adam T. Moody.
System Noise Revisited: Enabling Application Scalability
and Reproducibility with SMT.
IEEE International Parallel and Distributed
Processing Symposium, May 2016.
- SC 2015
- Wei Wang and Edgar A. Leon.
Evaluating DVFS and Concurrency Throttling on IBM's Power8
Architecture.
IEEE/ACM International Conference for High
Performance Computing, Networking, Storage and Analysis;
Research Poster, November 2015.
- SC 2015
- Bo Li, Edgar A. Leon, and Kirk W. Cameron.
Characterizing Memory Throttling Using the Roofline
Model.
IEEE/ACM International Conference for High
Performance Computing, Networking, Storage and Analysis;
Research Poster, November 2015.
- MEMSYS 2015
- ChunYi Su, Edgar A. Leon, Gabriel Loh, David Roberts,
Kirk W. Cameron, Dimitrios S. Nikolopoulos, and Bronis R. de
Supinski.
HpMC: An Energy-aware Management System of Multi-level
Memory Architectures.
International Symposium on Memory Systems,
October 2015.
- Cluster 2015
- Edgar A. Leon, Ian Karlin, and Ryan E. Grant.
Optimizing Explicit Hydrodynamics for Power, Energy, and
Performance.
IEEE International Conference on Cluster
Computing, September 2015. Best paper award.
- ModSim 2015
- Ian Karlin, Edgar A. Leon, Veselin Dobrev,
Christopher Earl, and Tzanio Kolev.
Performance, Power, and Resilience Tradeoffs in
Algorithmic Design.
DOE Workshop on Modeling & Simulation of Systems
and Applications, August 2015.
- WoC 2015
- Judicael A. Zounmevo, Swann Perarnau, Kamil Iskra, Kazutomo Yoshii,
Roberto Gioiosa, Brian C. Van Essen, Maya B. Gokhale,
and Edgar A. Leon.
A Container-Based Approach to OS Specialization for
Exascale Computing.
IEEE International Workshop on Container Technologies
and Container Clouds, March 2015.
- SC 2014
- Bo Li and Edgar A. Leon.
Power shifting opportunities on BG/Q using memory
throttling.
IEEE/ACM International Conference for High Performance
Computing, Networking, Storage and Analysis; Research Poster,
November 2014.
- SC 2014
- Eli Rosenthal and Edgar A. Leon.
Characterizing application sensitivity to network
performance.
IEEE/ACM International Conference for High Performance
Computing, Networking, Storage and Analysis; Research
Poster, November 2014.
- HotPower 2014
- Bo Li and Edgar A. Leon.
Memory throttling on BG/Q: A case study with explicit
hydrodynamics.
USENIX Workshop on Power-Aware Computing and Systems, October
2014.
- OSDI 2014
- Judicael A. Zounmevo, Kamil Iskra, Kazutomo Yoshii, Roberto
Gioiosa, Brian C. Van Essen, Maya B. Gokhale, and Edgar
A. Leon.
A single-kernel approach to OS specialization and node
resource partitioning for exascale computing.
USENIX Symposium on Operating Systems Design and
Implementation; Research Poster, October 2014.
- ModSim 2014
- Dong Li, Edgar A. Leon, and Bronis R. de Supinski.
Adaptive parallelism: Integrated performance, power, and
resilience modeling.
U.S. DOE Workshop on Modeling & Simulation of Systems and
Applications, August 2014.
- HPPAC 2014
- Edgar A. Leon and Ian Karlin.
Characterizing the impact of program optimizations on
power and energy for explicit hydrodynamics.
IEEE Workshop on High-Performance, Power-Aware Computing, May
2014.
- ScalA 2013
- Ignacio Laguna, Edgar A. Leon, Martin Schulz, and
Mark Stephenson.
A study of application-level recovery methods for
transient network faults.
ACM/IEEE Workshop on Latest Advances in Scalable Algorithms
for Large-Scale Systems, November 2013.
- SC 2013
- Eli Rosenthal, Edgar A. Leon, and Adam T. Moody.
Mitigating system noise with simultaneous
multi-threading.
ACM/IEEE International Conference for High Performance
Computing, Networking, Storage and Analysis; Research
Poster, November 2013.
- IISWC 2012
- ChunYi Su, Dong Li, Dimitrios S. Nikolopoulos, Kirk
W. Cameron, Bronis R. de Supinski, and Edgar A. Leon.
Model-based, memory-centric performance and power
optimization on NUMA multiprocessors.
IEEE International Symposium on Workload Characterization,
November 2012.
- SC 2012
- ChunYi Su, Edgar A. Leon, and Bronis de Supinski.
Optimizing pF3D using model-based, dynamic
parallelism.
ACM/IEEE International Conference for High Performance
Computing, Networking, Storage and Analysis; ACM Student
Research Competition, November 2012.
- SC 2012
- Ian Karlin, Jim McGraw, Esthela Gallardo, Jeff
Keasler, Edgar A. Leon, and Bert Still.
Memory and parallelism exploration using the LULESH proxy
application.
ACM/IEEE International Conference for High Performance
Computing, Networking, Storage and Analysis; Research Poster,
November 2012.
- HPDC 2011
- Edgar A. Leon, Rolf Riesen, Kurt B. Ferreira, and Arthur
B. Maccabe.
Cache injection for parallel applications.
ACM International Symposium on High-Performance
Parallel and Distributed Computing, June 2011.
- SC 2009
- Edgar A. Leon, Rolf Riesen, Arthur B. Maccabe, and Patrick
G. Bridges.
Instruction-level simulation of a cluster at scale.
ACM/IEEE International Conference on High-Performance
Computing, Networking, Storage, and Analysis, November
2009.
- Hot Interconnects 2007
- Edgar A. Leon, Kurt B. Ferreira, and Arthur
B. Maccabe.
Reducing the impact of the memory wall for I/O using cache injection.
IEEE Symposium on High-Performance Interconnects, August
2007.
- Euro MPI 2003
- Edgar A. Leon, Arthur B. Maccabe, and Ron
Brightwell.
An MPI tool to measure application sensitivity to
variation in communication parameters.
Recent Advances in Parallel Virtual Machine and Message
Passing Interface, Lecture Notes in Computer
Science. Springer-Verlag Berlin Heidelberg, September
2003.
- HSLN 2002
- Edgar A. Leon, Arthur B. Maccabe, and Ron Brightwell.
Instrumenting LogP parameters in GM: Implementation and validation.
IEEE Workshop on High-Speed Local Networks, November
2002.