The Architecture and Compilers Reading Group provides interested CIS graduate students an opportunity to read and discuss recent research in the field. All Penn students are welcome to subscribe to our mailing list, attend, suggest readings, and participate in the weekly discussions.

Weekly Meetings: Mondays, 11:00am-11.59am, Levine 512

Next Meeting:
Sept 26 EnerJ: Approximate Types for Safe and General Low Power Computation - PLDI 2011

• Sept 19: Web Search Using Mobile Cores: Quantifying and Mitigating the Price of Efficiency, ISCA 2010,
• Debunking the 100X GPU vs CPU myth: an evaluation of computing on CPU and GPU
• Pocket Cloudlets -- ASPLOS 2011
• EnerJ: Approximate Data Types for Safe and General Low-Power Computation -- PLDI 2011
• Exploiting Half-wits: Smarter Storage for Low-Power Devices, FAST'11
• MAUI: Making Smartphones Last Longer with Code Offload, MobiSys'10
• Debunking the 100X GPU vs. CPU myth: an evaluation of throughput computing on CPU and GPU, ISCA'10
• From Microprocessors to Nanostores: Rethinking Data-Centric Systems, IEEE Computer Jan'11.
• Cycles, Cells and Platters: An Empirical Analysis of Hardware Failures on a Million Consumer PCs , EuroSys'11.
• Google-Wide Profiling: A Continuous Profiling Infrastructure For Data Centers, IEEE Micro, Jul 2010.
• Power-Aware Computing with Dynamic Knobs, ASPLOS'11.
• Cohesion: A Hybrid Memory Model for Accelerators, ISCA 2010.
• A Survey of Hardware Trojan Taxonomy and Detection, IEEE Computer 2010.

Please send suggestions and opinions to Santosh Nagarakatte.

Decide on the paper by Wednesday at the latest (for Monday meetings), update this webpage so that it is current, then email the group list with paper name and link.

Usually, a good starting point for discussion is what problem the authors were trying to solve and how they propose to solve it. From there, have some specific discussion points in case discussion flags. There shouldn't be a need for much summary; everyone should have read the paper and will ask if they are confused about something.

Read the paper. This link provides some details on how to read a research paper. Try to give at least an hour to it, if not more; we are planning on reading papers that are worth spending time on. Think about the following questions: What problem are the authors trying to solve? How do they propose to solve it? Also note anything you are confused by, disagree with, or have some comment on, and bring it up in discussion.