In general-purpose graphics processing unit (GPGPU) computing, data is processed by concurrent threads executing the same function. This model, dubbed single-instruction/multiple-thread (SIMT), requires programmers to coordinate the synchronous execution of similar operations across thousands of data elements. To alleviate this programmer burden, Gaster and Howes outlined the channel abstraction, which facilitates dynamically aggregating asynchronously produced fine-grain work into coarser-grain tasks. However, no practical implementation has been proposed.
Drop in to see part two of student demonstrations of software projects created in CS 638 (Associate Professor Ben Liblit's Undergraduate Topics in Computing course). On this day, the last four groups will demonstrate.
The day's projects are: American Ultimate Disc League (AUDL) Mobile Application, Perfect Deck of Cards, My Fridge and Glassroom (Google Glass in the Classroom).
Drop in to see student demonstrations of software projects created in CS 638 (Associate Professor Ben Liblit's Undergraduate Topics in Computing course). On this day, the first four out of eight groups will show their work to the public; come back Thursday during the same time period for the remaining four.
Tuesday's projects are: Dewey Decibel Database (D3), Minecraft RPG Expansion, Online Board Game and Wisconsin Land Cover Mapping Field Validation Tool.
Storage technology is changing rapidly. New hardware promises unprecedently low latency, while traditional disks are becoming increasingly unreliable in an attempt to meet capacity demands. Meanwhile, cloud models, be they fully public, private, or a hybrid of the two, have made storage both more flexible and more fragmented, a trend accentuated by the proliferation of data into mobile clients.