Electrical Engineering
Resume:
Elkin E. Garcia
San Jose, CA 95124
Phone: +1-818-***-****
*******@****.***
http://www.capsl.udel.edu/~egarcia
Interests
Top expert in machine learning and parallel computing with strong interests in classification, clustering,
regression, parallel programming, development of high performance applications and
optimization/modeling of parallel algorithms looking for a challenging position in the industry.
Education
Ph.D. in Electrical & Computer Engineering. Mar. 2014 (expected)
University of Delaware, Newark, DE.
Thesis Topic: High Performance and Energy Efficiency on many-core Architectures.
Advisor: Prof. Guang R. Gao. GPA 3.895/4.000.
M.Sc. in Electrical & Computer Engineering. Dec. 2005
Universidad de los Andes, Bogota, COL.
Thesis Topic: Machine Learning and Signal Processing applied to Pattern Recognition.
GPA 4.56/5.00. Class Rank: 1 out of 8.
B.S. in Electrical Engineering. Jun. 2003
Pontificia Universidad Javeriana, Bogota, COL.
GPA 4.32/5.00. Class Rank: 1 out of 50+.
Skill Set
High proficiency in C/C++, OpenMP, Pthreads, MPI, Shell Scripting, CUDA, Matlab.
Others: Assembly, VHDL, OpenCL
Experience
Research Assistant, University of Delaware, Newark, DE. Jan. 2009 – Present
Research and development in the area of Parallel Computer Architecture under the supervision of Prof.
Guang R. Gao. Work in the Electrical and Computer Engineering Department as part of the Computer
Architecture and Parallel System Laboratory (CAPSL). Most relevant projects/activities are:
Drive the implementation, evaluation, optimization and tuning of several algorithms (e.g.
Linear algebra, FFT, stencil) to drastically increase the performance and energy efficiency on
many-core architectures. Major focus on tiling, scheduling and communication.
Co-lead the design, implementation and evaluation of scalable fine-grain multi-threading
energy-aware data-driven parallel execution models and runtime system.
Lead the design and implementation of energy and performance modeling of many-core
architectures with resource constraints.
Graduate Intern Research Scientist, Intel Corp., Santa Clara, CA Oct. 2013 - Jan. 2014
Research and development of dynamic runtime adaptations to manage tradeoffs between Performance
and Energy consumption in fine grain, event-driven runtime systems. Work under the Programming
Systems Labs (PSL) part of Intel Labs in collaboration with Rice University. Most relevant activities are:
Develop and optimize scalable/high performance/energy efficient algorithms. Particularly, I
have been using the Open Community Runtime (OCR) as SW test platform.
Enable monitoring of energy and performance metrics using core and uncore hardware counters
to feed adaptation policies at application and runtime system level.
Design, prototype and test an adaptive runtime system using policies to dynamically control
energy consumption and performance.
Research Intern, Intel Corporation, Hillsboro, OR Jun. 2011 - Jan. 2012
Developed FPGA technologies attached to Intel Processors. Worked in the Intel Architecture Group (IAG)
as part of Accelerator Technology Path finding (ATP). The project was:
Platform Support Package for an FPGA accelerator attached to a cache coherent system fabric:
Developed a framework for increasing programmability and productivity into the Intel QPI-based FPGA
accelerator Platform (QAP). Investigated and adapted tools such as C to Gates compilers for automatic
generation of software libraries and RTL infrastructure in QAP.
Research Intern, University of Delaware, Newark, DE Jun. 2008 - Dec. 2008
Developed parallel applications using a wide set of tools and platforms as a member of the Computer
Architecture and Parallel System Laboratory (CAPSL). Focused on:
Multi-platform APIs for Parallel Programming such as OpenMP, POSIX threads and MPI.
Multi-core and Many-core Architectures such as x86, Cell Broad Engine and Cyclops-64.
Instructor, Universidad de los Andes, Bogota, COL. Jun. 2006 - Aug. 2008
Supervised senior projects on microelectronics, electronic design, signal processing and intelligent
systems. Member of the microelectronic group (CMUA) in charge of the Electronic System Design Area.
Most relevant projects were:
Beam former Design on FPGA: Developed a transmitter beam former on FPGA.
Machine Learning Techniques Applied to Bioengineering: Designed a computer mouse control
system for disabled using Boosting.
Research Assistant, Universidad de los Andes, Bogota, COL. Jan. 2005 - Dec. 2005
Research and development in the area of Machine Learning and Signal Processing. Worked in the
Electrical Department with the microelectronic group (CMUA) under the supervision of Prof. Fernando
Lozano. Relevant projects were:
Non Stationary Time Series Prediction with Support Vector Machines (Jan. - Jun.): Developed a
new SVM algorithm to predict non stationary time series based on the estimation of stationary
series.
Boosting Techniques with Support Vector Machines (Jul. - Dec.): Implemented a faster SVM
training algorithm by combining several SVM weak learners with Adaboost
Selected Publications
1. Daniel Orozco, Elkin Garcia, Rishi Khan, Kelly Livingston and Guang Gao, “Toward High
Throughput Algorithms on Many Core Architectures”, TACO 2012.
2. Elkin Garcia, Daniel Orozco, Robert Pavel and Guang Gao, “A discussion in favor of Dynamic
Scheduling for regular applications in Many-core Architectures”, MTAAP - IPDPS 2012.
3. Elkin Garcia, Daniel Orozco and Guang Gao, Energy efficient tiling on a Many-Core Architecture,
MULTIPROG - HiPEAC 2011. Best Paper Award.
4. Daniel Orozco, Elkin Garcia and Guang Gao, Locality Optimization of Stencil Applications using
Data Dependency Graphs, LCPC 2010.
5. Elkin Garcia and Fernando Lozano, “Boosting Support Vector Machines”, MLDM 2007.
6. Pedro Rangel, Fernando Lozano and Elkin Garcia, “Boosting of Support Vector Machines with
application to editing”, ICMLA 2005.
Selected Awards & Honors
University Graduate Fellowship. 2012 - 2013.
Computer System and Networking Graduate Faculty Award. 2011 and 2012.
Best Paper Award MULTIPROG-HiPEAC. 2011
Best Master Thesis in Electrical and Computer Engineering. 2006.
Outstanding Student Distinction within Class'03 Graduates. 2003.
Thesis Project Nomination for Best Project in Electrical Engineering. 2003.
Andres Bello Fellowship. 1998 - 2002.
Father Gerardo Remolina Scholarship. 2000 - 2001.
Contact this candidate