2007 — 2010 |
Sohoni, Sohum |
N/AActivity Code Description: No activity code was retrieved: click on the grant title for more information |
Csr-Pdos: Out-of-Context Prefetching For L2 Caches @ Oklahoma State University
On a multi-tasked system, the CPU cycles through a number of processes every few milliseconds. Some of the processes are memory-intensive, but most are non-intensive support processes and other user programs such as text-editors, email clients, and instant messengers. A typical system cycles through a few memory-intensive processes, each of which replaces several cache blocks from the level-2 cache from other processes. This results in a large number of demand misses for the L2, and ties up the memory subsystem preventing effective prefetching.
This project investigates a possible solution to this problem by modifying the OS to prefetch to the L2 out of context, i.e. when the CPU is executing a non memory-intensive process. (ideally, just before the memory-intensive process will be brought in for execution). Thus, when this process resumes execution, it will have low L2 cache misses which in turn will keep the memory-bus free in order to prefetch even more blocks. Out-of-context prefetching aims to shift from a predominantly demand-fetched paradigm to a predominantly prefetched paradigm.
In conjunction with accurate predictors for prefetching, this work has the potential to greatly mitigate the performance CPU-memory gap problem on a system-wide level. The expected outcome of the project is a Linux implementation of out-of-context prefetching on a full-system simulator, and a website that documents all the assumptions, the source code, and extensive project results beyond those published in the corresponding research literature. All researchers, operating system designers, and processor manufacturers will have access to the website and the research findings to freely incorporate them as they wish.
|
0.939 |
2011 — 2015 |
Sohoni, Sohum Zhu, Lan (co-PI) [⬀] Komanduri, Ranga (co-PI) [⬀] Brunson, Dana Singh, Raman (co-PI) [⬀] Hoyt, Peter |
N/AActivity Code Description: No activity code was retrieved: click on the grant title for more information |
Mri: Acquisition of a High Performance Compute Cluster For Multidisciplinary Research @ Oklahoma State University
Under this Major Research Instrumentation (MRI) project, Oklahoma State University High Performance Computing Center (OSUHPCC) will acquire, deploy and maintain an HPC cluster supercomputer, to be named Cowboy, that will support computing-intensive research and research training across a broad variety of Science, Technology, Engineering and Mathematics (STEM) disciplines. As a campus-wide shared resource, Cowboy will be available not only to all of OSU's faculty, staff, postdocs, graduate students and undergraduates, but to researchers across Oklahoma.
Many areas of Computational and Data-Enabled Science and Engineering (CDESE) research will be facilitated by the proposed system by collaborating research teams with an expected doubling of the number of users every 12 months based on experience to date. Projects include: mechanics of granular materials and fracture simulations in nuclear clad materials; discovery genes for canine hip dysplasia; improvements to performance per Watt of many-core systems; transcriptional profiling of determination events in adult and embryonic murine stem cell lines; genomic, metagenomic and proteomic approaches to decipher host- pathogen interactions, complex carbohydrate metabolism and cellulosic bioenergy; modeling geophysical fluids; computational chemistry; simulations of nanostructured materials; simulation of the growth of carbon nanotubes; simulating gas-phase and condensed-phase materials; seismic characterization of surface geology; computational optimization; optimal error-control coding and compressive sensing techniques; charge, spin and heat transport; molecular phylogeny of the Asteraceae; evolutionary genetics of morphological diversification and domestication in grasses (Poaceae) and mustards (Brassicaceae); robust electromagnetic field testing; phylogenomic analyses of the extremophile red alga Galdieria sulphuraria; integrating data in evolving social networks; computational and combinatorial methods in commutative algebra; microbes associated with non-cultivated and cultivated plants; intensity-modulated radiation therapy planning software; and phylogenomics of milkweeds (Asclepias, Apocynaceae).
Oklahoma users are at 24 institutions, including 11 of Oklahoma's 13 public universities. The Oklahoma Supercomputing Symposium (MRI PI Brunson is 2011 Conference Co-Chair), the only event of its kind held annually in an EPSCoR jurisdiction, in 8 years has had over 2000 attendees from 92 academic institutions in 24 states and Puerto Rico (including 33 in Oklahoma and 26 in 12 other EPSCoR jurisdictions), 97 private companies, 31 government agencies and 14 nonprofits. This project includes seven women senior personnel who serve as role models and are creating interest and excitement about computational science and engineering. This project also includes participation of two researchers at Langston University, Oklahoma's sole Historically Black University. Research projects include 80 graduate students and 29 undergraduates, including 29 women, an African American, three Native Americans and three Hispanics.
|
0.939 |
2011 — 2015 |
Sohoni, Sohum Cho, Yoonjung (co-PI) [⬀] Damron, Rebecca (co-PI) [⬀] Kearney, Kerri (co-PI) [⬀] |
N/AActivity Code Description: No activity code was retrieved: click on the grant title for more information |
Rigee: Progressive Learning Platform For Computer Engineering @ Oklahoma State University
The long-term vision for this Research Initiation Grant in Engineering Education is to transform the way computer engineering is taught by establishing the computer engineering curriculum around a carefully designed learning platform. This engineering education project involves research and faculty collaborators from Computer Engineering, the English Department, and the College of Education. It aims to introduce the learning platform system in various computer engineering courses and perform qualitative and quantitative studies to gauge the impact on student learning. The project will generate new knowledge on how students learn, retain, and transfer engineering concepts in a community of practice. It will also test whether students? fixed mindsets can be changed to growth mindsets.
This research initiation project will provide a vehicle for the PI to explore PLP?s impact on student learning through well-known frameworks and assessment models. The impact on the PI and his graduate students will be significant, as they will learn the intricacies of quantitative and qualitative research in education. Undergraduate students will benefit through a well-designed educational tool. Collaborations will be formed that will lead to future projects where experts from the social and educational sciences will work with engineering faculty to perform rigorous research on student learning.
|
0.943 |