David O'Hallaron

David O'Hallaron

Professor of Computer Science And Electrical And Computer Engineering

Office Room 7517 Gates and Hillman Centers

Email droh@andrew.cmu.edu

Phone (412) 268-8199

Department
Computer Science Department

Website
http://www.cs.cmu.edu/~droh

Administrative Support Person
Angy Malloy

Research Interests
Computer Architecture

Research Statement

I work in the area of computer systems. My interests within computer systems are quite broad, including online education, cloud computing, and scientific computing. My main interest these days is the phenomena surrounding autograding, that is, programs evaluating other programs.

The CMU Autolab group is developing a new cloud-based service that teachers around the world can use to offer programming labs for their computer classes. The service is based on the notion of autograding, that is, programs evaluating other programs. Our vision is that teachers can select the labs for their classes from a repository of high-quality labs written by other teachers and students. An author whose lab is adopted for a class receives community recognition, in the form of a public adoptions page, and possibly even a small royalty. Each time a student hands in their work for credit, the service spins up a new VM and autogrades the student's work in this new VM. The scores are displayed, anonymized, on a realtime scoreboard that is visible to everyone in the class.

Our aim with this work is to improve the quality of computer science education worldwide by providing a way for teachers to share their good labs with each other. This sharing will enable good teachers to get credit for their work from their peers. The assignment of credit, in turn, will help foster the birth of a new reputation-based community for teachers. Researchers have enjoyed this kind of community for years, but teachers have never developed one, working for the most part in isolation, with little opportunity to learn and benefit from each other.

Students who are passionate about education would enjoy working with us. Drop by any time if you want to chat.

Recent Publications

Campbell R, Gupta I, Heath M, Ko SY, Kozuch M, Kunze M, Kwan T, Lai K, Lee HY, Lyons M, Milojicic D, O'Hallaron D, Soh YC ( 2020 ) Workshop on Hot Topics in Cloud Computing, HotCloud 2009, Open cirrus<sup>TM</sup> cloud computing testbed: Federated data centers for open source systems and services research

Campbell R, Gupta I, Heath M, Ko SY, Kozuch M, Kunze M, Kwan T, Lai K, Lee HY, Lyons M, Milojicic D, O'Hallaron D, Soh YC ( 2020 ) Workshop on Hot Topics in Cloud Computing, HotCloud 2009, Open cirrus<sup>TM</sup> cloud computing testbed: Federated data centers for open source systems and services research

Parlante N, Zelenski J, Zingaro D, Wayne K, O'Hallaron D, Guerin JT, Davies S, Kurmas Z, Debby K ( 2012 ) SIGCSE'12 - Proceedings of the 43rd ACM Technical Symposium on Computer Science Education, Nifty assignments , Page(s): 475 - 476

David O'Hallaron ( 2010 ) Computer, Open cirrus: A global cloud computing testbed , Vol.: 43 Issue: ( 4 ) , Page(s): 35- 43 .

Recent Awards

Benjamin Richard Teare Teaching Award, College of Engineering (CIT) | 2017 |

HPC Analytics Challenge, Super Computing | 2006 | Best Paper Award
(with Hongfeng Yu, UC-Davis), Tiankai Tu, Jacobo Bielak, Omar Ghat- tas (University of Texas at Austin), Julio Lopez, Kwan-Liu Ma (UC-Davis), Leonardo Ramirez-Guzman, Nathan Stone (PSC), Ricardo Taborda-Rios, John Urbanic (PSC)

Gordon Bell Award for Special Achievement | 2003 | Faculty Research Award
High Resolution Forward and Inverse Earthquake Modeling on Terascale Computers. With with Jacobo Bielak, Omar Ghattas, and our students The milestone calculations for the award included: • The generation of a record unstructured hex mesh (3.7 billion elements, 4 billion nodes) • The largest unstructured mesh wave propagation simulation (900 million elements, 3.2 billion DOF) • The largest acoustic wave propagation inverse problem (17 million inversion parameters, 70 billion total unknowns) • The largest elastic wave propagation inverse problem (275,000 inversion parameters, something like a billion total unknowns). Program 183.EQUAKE (earthquake ground motion modeling) selected by SPEC (Standard Performance Evaluation Corporation) as one of 14 floating point benchmarks in the SPEC CPU2000 benchmark suite. Parallelized versions of 183.EQUAKE, called 320.EQUAKE M and 321.EQUAKE L selected for inclusion in the SPEC OMPM2001 and OMPL2001 benchmark suites for evaluating shared-memory multiprocessor system performance.

Allen Newell Faculty Award for Reseach Excellence | 1998 | Faculty Research Award
The Quake Project