Welcome to ICSE@UM
The Institute for Computational Science & Engineering (ICSE) is a new initiative at the University of Michigan. ICSE creates opportunities and leverages synergies in computational science research and education across UM and operates in collaboration with the Office of Research Cyberinfrastructure (ORCI), Information Technology Services (ITS), and the various schools and colleges at UM to foster a culture and community for scientific computing through the development and nuturing of new activities, services and resources for faculty, staff and students. ICSE is committed to providing a comprehensive, state-of-the-art, scientific computing environment in support of research excellence at the University of Michigan.
Announcements & Opportunities
ICSE Associate Director position posted
ICSE seeks an Associate Director to oversee all day-to-day operations; develop and implement activities and resources for faculty, staff and students; organize activities and events; participate in development, fundraising and reporting; and implement the vision and goals of the Institute. more info...
Faculty Positions in Petascale Computing
The Departments of Materials Science & Engineering, Atmospheric, Oceanic and Space Sciences, Civil and Environmental Engineering, Mathematics, and Electrical Engineering & Computer Science at the University of Michigan jointly seek applicants at the assistant professor level in the area of petascale computing. Apply now!
Great Lakes Consortium for Petascale Computation. UM is a founding partner in the GLCPC. The Consortium is a key element of the Blue Waters Project at NCSA, and facilitates the effective and widespread use of petascale computing through the development of new computing software, applications, and technologies. To get involved in the GLCPC or Blue Waters, contact sglotzer@umich.edu.
$$$ for Scientific Computing. Funding for scientific computing is on the rise! Click here to see a list of new solicitations in CS&E from federal agencies, including several new programs at the National Science Foundation.
ICSE seeks an Associate Director to oversee all day-to-day operations; develop and implement activities and resources for faculty, staff and students; organize activities and events; participate in development, fundraising and reporting; and implement the vision and goals of the Institute. more info...
Faculty Positions in Petascale Computing
The Departments of Materials Science & Engineering, Atmospheric, Oceanic and Space Sciences, Civil and Environmental Engineering, Mathematics, and Electrical Engineering & Computer Science at the University of Michigan jointly seek applicants at the assistant professor level in the area of petascale computing. Apply now!
Great Lakes Consortium for Petascale Computation. UM is a founding partner in the GLCPC. The Consortium is a key element of the Blue Waters Project at NCSA, and facilitates the effective and widespread use of petascale computing through the development of new computing software, applications, and technologies. To get involved in the GLCPC or Blue Waters, contact sglotzer@umich.edu.
$$$ for Scientific Computing. Funding for scientific computing is on the rise! Click here to see a list of new solicitations in CS&E from federal agencies, including several new programs at the National Science Foundation.
CodeBlue. Are you a UM researcher with a new code development project? Use our new Trak server at http://codeblue.umich.edu to create a free Trak site you control for your open-source or private code development projects. For more information, click here.
Funding Opportunities @ ICSE. Beginning Fall 2010, ICSE will provide competitive funding to the UM CoE community for undergraduate scholarships, graduate fellowships, visitor fellowships, seed funds for collaborative, high-risk research, and funds to support ICSE workshops and symposia in scientific computing
Funding Opportunities @ ICSE. Beginning Fall 2010, ICSE will provide competitive funding to the UM CoE community for undergraduate scholarships, graduate fellowships, visitor fellowships, seed funds for collaborative, high-risk research, and funds to support ICSE workshops and symposia in scientific computing
Upcoming Events
ICSE Seminar - Computing with Chebfun
Nick Trefethen, Professor of Numerical Analysis, Oxford University
Wednesday March 9, 12-1pm, 1005 DOW
Anyone who uses Matlab will enjoy Chebfun, a freely-available package that overloads Matlab's familiar operations on vectors to apply to functions of a continuous variable. Thus in Matlab, SUM(f) adds up the entries of a vector, but in Chebfun, it computes the integral of a function. MAX(f) finds the maximum, ROOTS(f) finds the roots, and so on. Under the hood, everything is based on piecewise expansions in Chebyshev polynomials, but in principle you don't need to know that.
This talk will be a hands-on demonstration, and attendees are welcome to bring a laptop running Matlab if they want try Chebfun on the spot. Just as Chebfun turns vectors into functions, it turns matrices into operators, notably differential and integral operators. One of our proudest achievements is that we've overloaded Matlab's backslash command to solve linear and nonlinear differential equations, typically to close to machine precision in a fraction of a second.
Wednesday March 9, 12-1pm, 1005 DOW
Anyone who uses Matlab will enjoy Chebfun, a freely-available package that overloads Matlab's familiar operations on vectors to apply to functions of a continuous variable. Thus in Matlab, SUM(f) adds up the entries of a vector, but in Chebfun, it computes the integral of a function. MAX(f) finds the maximum, ROOTS(f) finds the roots, and so on. Under the hood, everything is based on piecewise expansions in Chebyshev polynomials, but in principle you don't need to know that.
This talk will be a hands-on demonstration, and attendees are welcome to bring a laptop running Matlab if they want try Chebfun on the spot. Just as Chebfun turns vectors into functions, it turns matrices into operators, notably differential and integral operators. One of our proudest achievements is that we've overloaded Matlab's backslash command to solve linear and nonlinear differential equations, typically to close to machine precision in a fraction of a second.
Virtual School of Computational Science & Engineering. Want to learn how to use graphics processors for scientific computing? Scale your parallel code to tens of thousands of CPU cores? Deal with ginormous datasets? The VSCSE will again be offering multiple courses for the summer of 2011. The course materials from 2010 - including video, audio, slides, labs, tutorials and more - are available online at the VSCSE hub!
