01.07.2014 Views

Download PDF (9.49MB) - the School of Engineering and Applied ...

Download PDF (9.49MB) - the School of Engineering and Applied ...

Download PDF (9.49MB) - the School of Engineering and Applied ...

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

WWW.SEAS.UPENN.EDU<br />

SCHOOL NEWS<br />

New Faculty<br />

Igor Bargatin<br />

Class <strong>of</strong> 1965 Term Assistant Pr<strong>of</strong>essor <strong>of</strong><br />

Mechanical <strong>Engineering</strong> <strong>and</strong> <strong>Applied</strong> Mechanics<br />

Ph.D. in 2008 in Physics <strong>and</strong> Electrical <strong>Engineering</strong>,<br />

California Institute <strong>of</strong> Technology<br />

Postdoctoral Fellow at Stanford University<br />

Dr. Bargatin’s research interests are focused on micro- <strong>and</strong><br />

nanoelectromechanical systems (MEMS/NEMS) for new<br />

applications in energy conversion, optics <strong>and</strong> smart<br />

materials. He applies his knowledge <strong>of</strong> physics <strong>and</strong><br />

engineering disciplines to design <strong>and</strong> test new types <strong>of</strong><br />

devices, such as micr<strong>of</strong>abricated <strong>the</strong>rmionic energy<br />

converters, which convert heat directly to electricity at very<br />

high temperatures by literally boiling electrons <strong>of</strong>f a surface<br />

<strong>and</strong> using <strong>the</strong>m as a “working fluid” in a heat engine. The<br />

Bargatin Group engages in interdisciplinary research at <strong>the</strong><br />

intersections <strong>of</strong> mechanical engineering, electrical<br />

engineering, materials science <strong>and</strong> applied physics.<br />

Joseph Devietti<br />

Assistant Pr<strong>of</strong>essor <strong>of</strong> Computer <strong>and</strong> Information Science<br />

Ph.D. in 2012 in Computer Science & <strong>Engineering</strong>,<br />

University <strong>of</strong> Washington<br />

Sequential programming has become steadily safer <strong>and</strong> easier<br />

over time, but parallel programming, despite its importance<br />

with today’s ubiquitous multicore processors, has made much<br />

more limited progress. Dr. Devietti works on identifying new<br />

safety properties for parallel programs, precisely defining <strong>the</strong>ir<br />

guarantees <strong>and</strong> exploring efficient implementations employing<br />

a range <strong>of</strong> hardware <strong>and</strong> s<strong>of</strong>tware mechanisms. A recent<br />

project demonstrated how to automatically <strong>and</strong> precisely<br />

identify concurrency bugs in parallel programs. Dr. Devietti<br />

has also conducted research showing how to simplify parallel<br />

programming by automatically eliminating <strong>the</strong> nondeterminism<br />

that arises from current multicore architectures <strong>and</strong><br />

programming models. With nondeterminism removed, parallel<br />

programs behave like conventional sequential programs that<br />

are much easier to write, debug <strong>and</strong> underst<strong>and</strong>.<br />

SPRING 2013 n 34

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!