Evan Speight Email & Phone Number

Austin, Tx, Us

Austin, Tx, Us

Austin, Texas Metropolitan Area

Austin, Texas, United States

Austin, Texas Area

Building on my twenty years of experience in high performance computing, I co-founded Radyalis in November 2013 to bring to the market a set of treatment planning algorithms, real-time Monte Carlo dose calculations, and highly optimized treatment planning for the proton therapy cancer treatment space. We are collaborating with partners at premier proton and photon therapy research institutions to explore improved methods for treatment that greatly reduce cost and long delays between imaging and first treatment. Our current offering is a fast Monte-Carol simulation tool that advances by 5000X the current state-of-the-art solution on commodity hardware. Our current offering is a Monte Carlo secondary dose calculator available in the marketplace in the Delta4 Insight Offering by ScandiDos, a world leader in dosimetry and QA.

Austin, Texas Area

While at IBM I worked on many different projects applying HPC concepts and systems to other domains. From 2/2011 to 2/2012 I served as manager of the Workload Optimized System group in the Austin Research Lab. My group focused on the development of architectural features and system software that allow inexpensive customization for each potential customer's needs. Specifically, we improved the performance of "big data" business analytics software packages through performance modeling, analysis, and system design. Our innovations on solutions to support packages such as Big Insights allowed customers to analyze petabytes of semi-structured or unstructured data quickly and accurately.From 2003 to 2011 I worked on the PERCS project at IBM, developing a POWER 7-based supercomputer under the HPCS program sponsored by DARPA. I researched key areas of computer architecture that appear in the POWER7 processor, distributed systems as embodied in the high-end P775 offering from IBM, and modeling the performance of such large-scale systems. I have 44 patents relating to my work on PERCS and other projects and have published 32 papers, including publications in top conferences such as ISCA, HPCA, and ICS.

At Cornell, I was a tenure-track university professor in the School of Electrical Engineering. As such, I mentored graduate students, conducted research, obtained funding from the National Science Foundation and other agencies, published papers in scholarly journals and conference proceedings, and taught both undergraduate and graduate classes, winning an award for "Outstanding Undergraduate Professor" during my second year.

Lead a team of researchers in the development and implementation of Brazos, a parallel programming environment for use on networks of PC workstations and servers running Windows NT; implemented mixed shared-memory and message-passing programming support; experimented with emerging user-level network architectures; designed and implemented software coherence protocols; researched the effects of multiprogramming, thread migration, and fault tolerance in the development of cluster-based runtime systems.Previously at Rice I examined techniques for performance tuning of parallel programs on hardware shared memory systems; developed ParaView, a graphical tool based on MOTIF and Xt for uncovering program inefficiencies in shared memory programs; and wrote a shared memory parallel simulator based on YACSIM, an event-driven simulator used extensively in parallel research.

Developed the methodology for use in statistical fault grading on the 82360SL I/O subsystem in the Intel 386SL chip-set. Improved fault coverage through test vector refining. Created an extensive suite of analysis tools to cover the efficiency of all phases of the manufacturing process. Member of the team that completed and released the wafer-sort test program for the Intel 386SL chip set.

Erhlangen

Constructed electronic circuits to test components for a gradient amplifier to be used in a magnetic resonance imaging system. Wrote programs in Pascal to chart power losses throughout entire system.

Greater Memphis Area

Summer intern with nuclear medicine, imaging, and radiation treatment department.

Ph.D, Electrical And Computer Engineering

Thesis title: Efficient Runtime Support for Cluster-Based Distributed Shared Memory Multiprocessors Advisor: John K. Bennett

M.S, Electrical And Computer Engineering

Thesis: Paraview: Performance Debugging Through Visualization of Shared Data Advisor: John K. Bennett

B.S., Electrical Engineering

Activities and Societies: Stanford Overseas Program (Germany)