NSF CCF-1533644: XPS: FULL: FP: A Profile-Centric IDE for Science-Based Performance Engineering in the Cloud.
Scientists developing compute-intensive multicore applications find it difficult to parallelize their code, a problem that is exacerbated if they wish to take maximum advantage of the potential provided by cloud computing. Part of the problem is that bad codes cause the generation of incorrect hypotheses while reading, writing, and debugging code, wasting time, energy, and resources.
This research plans to meld advanced profiling methods for multithreaded programming with modern user-interface technology to produce a highly usable open-source integrated development environment (IDE) for the performance-engineering of multicore software applications in the cloud. The goal is to provide programmers with continuous profile data for scalability and other performance profiling relevant to parallel programming. They plan to embed an IDE into a profiling framework to produce a profile-centric IDE, continuously providing performance feedback so that developers can see and compare the results of recent runs of their program as they edit their code.
The project has the potential to enable science-based performance engineering of multicore applications in the cloud. The vast majority of computer users, not just expert computer scientists, will be able to develop highly efficient parallel software applications, broadly impacting every computing application in every walk of life. The software produced by this project will be made freely available to anyone on the World Wide Web using a liberal open-source license.
DOE/NNSA PSAAP-III: CESMIX: Center for the Exascale Simulation of Material Interfaces in Extreme Environments.
The Center for Exascale Simulation of Materials in Extreme Environments (CESMIX) seeks to advance the state-of-the-art in predictive simulation by connecting quantum and molecular simulations of materials with state-of-the-art programming languages, compiler technologies, and software performance engineering tools, underpinned by rigorous approaches to statistical inference and uncertainty quantification.
Our motivating problem is to predict the degradation of complex (disordered and multi-component) materials under extreme loading, inaccessible to direct experimental observation. This application represents a technology domain of intense current interest, and exemplifies an important class of scientific problems — involving material interfaces in extreme environments.
LANL Subcontract No. 531711: LLVM Extensions for Parallel Computing.
USAF-MIT AI Accelerator: Fast AI: Quick Development of Portable High-Performance AI Applications.
The AI revolution has been enabled by the availability of vast amounts of labeled data, novel algorithms, and computer performance. But long computer-in-the-loop development cycles inhibit humans from inventing and deploying creative AI solutions. Moreover, the end of Moore’s has curtailed the historical ability of semiconductor technology to deliver performance. AI performance increasingly relies on hardware architecture, software, and algorithms. The Fast AI project focuses on developing a foundation for quickly building AI solutions, enabling performance and portability on both modern and legacy hardware platforms. We innovate in the areas of programming languages, compiler technologies, comprehensive instrumentation, analytical productivity tools, and parallel algorithms.
MIT-IBM Watson AI Lab Core Project: xGraph: Accelerated and Explainable Graph Deep Learning with Applications to Financial Services.