Name: Jose M Monsalve Diaz
Pronouns: He/him/his
Biography:
Jose M Monsalve Diaz is a postdoctoral appointee at Argonne National Laboratory working on exploring innovative ideas for future computer architectures. He obtained his Ph.D. in Electrical and Computer Engineering from the University of Delaware in 2021. His work defined the Sequential Codelet Model, a program execution model for future parallel, heterogeneous, and distributed computers based on principles of Dataflow models of computation. Jose M Monsalve Diaz obtained his Master’s in Electrical and Computer Engineering from the University of Delaware in 2020 and his bachelor’s degree in Electronics Engineering from the Pontificia Universidad Javeriana in 2013. Throughout the years, he has worked as a research assistant of the CAPSL research group for Prof Guang. R. Gao and the CRPL research group for Prof. Sunita Chandrasekaran. His areas of interest are parallel computer architecture design, parallel computer systems, and parallel programming models. He has worked on validating and verifying OpenMP target offloading and with OpenACC programming targeting CPU and heterogeneous systems based on GPGPUs. Other projects also involved unconventional Dataflow-based programming models and computer architectures.
Institution/Lab: Argonne National Laboratory
Website: https://www.anl.gov/profile/jose-manuel-monsalve
SRP Collaboration Topic/Title: A learning experience of compilers, runtime systems and parallel programmin
Field or research area: Compilers, Runtimes, Parallel programming, OpenMP, LLVM, HPC
Please select all the topical areas that apply to your project:
Computer Science (i.e., architectures, compilers/languages, networks, workflow/edge, experiment automation, containers, neuromorphic computing, programming models, operating systems, sustainable software)
Brief Abstract:
High-Performance Computing is an exciting field that constantly pushes the limits of computation. HPC is a complex field. Not only is it necessary to have scientific knowledge of the applications, but also to have a deep understanding of how systems work. The more knowledge of the systems and tools, the higher the scientific impact could be. In this project, we seek to allow learning about compilers, runtime systems, and parallel programming. We aim to find a project that considers the needs of the current projects at ANL in addition to the professional development of the participants. We plan on using parallel programming models such as OpenMP, SYCL, OpenACC, and others. We explore innovation in their runtime systems and compilers. We will use the LLVM compiler infrastructure as one of the most important compilers in the area of HPC. By the end, we would like to encourage participants to 1) be part of the LLVM community, 2) better understand compilers, runtime systems, and parallel programming, 3) get involved in the area of HPC, and 4) create the necessary grounds for professional development.
Desired relevant skills, background, or interests:
The following is a list of topics we could work on. It is not a checklist of requirements, but a list of topics we can use to start the discussion. – Programming languages: – C, C++ or Fortran – Python – Parallel programming: – Threading libraries – OpenMP, OpenACC, SYCL, or similar – MPI, RPC, or similar – Understanding of parallel programming concepts – Understanding of compilers: – User-level understanding (e.g., compiler flags) – CS level understanding (e.g., compilation pipeline and compiler stages) – Developer-level understanding (e.g., LLVM development) – HPC: – Cluster infrastructure – HPC network interconnects – HPC applications
Other comments:
Do any special requirements apply? International OK
Other, specify:
Keywords:
HPC; Runtime Systems; Compilers; OpenMP
Lightning Talk Title: Compiler and Runtime Development in Hpc