This event is a three-day tutorial that provides an overview of techniques for optimizing I/O in serial and parallel applications.
- Organizer: Ritu Arora
| Event Information | Details |
|---|---|
| Event Name | Optimizing I/O |
| Registration Deadline | June 9, 2023 |
| Event Dates | June 12-14, 2023 |
| Website | https://forms.gle/7rWkR6r6ZdQbjFaA7 |
A tutorial on optimizing I/O in serial and parallel applications will be offered from 12:30 PM-2:30 PM CST on June 12-June 14, 2023. The tutorial will include a high-level overview of the general strategies for optimizing I/O for improving code performance and will cover topics and examples related to parallel filesystems (Lustre), MPI I/O, and I/O in the context of checkpointing and AI. Hands-on exercises and demos will be included in the tutorial.
Prerequisites: Familiarity with Linux, C/C++ programming and basic MPI. A brief overview of the related basic concepts will be included in the tutorial where needed. Additionally, participants will need to have accounts on the HPC platforms of their interest if they need to run any examples or hands-on exercises.
Target Audience
The target audience for this tutorial primarily consists of individuals involved in scientific software development, particularly those who do not have prior experience with parallel filesystems, MPI I/O, and application-level checkpointing. Additionally, AI application developers working with frameworks such as Tensorflow and PyTorch would also benefit from this tutorial. Specifically, those interested in learning about checkpointing as a solution for time-limits on job queues on high-performance computing (HPC) platforms and creating workflows with dependent jobs would find the tutorial particularly useful.
Topics/Agenda
The tutorial is expected to run for approximately two hours on each of the training days. It will focus on various topics, including strategies and practices for reducing inefficient I/O in both serial and parallel applications. Some specific areas of focus will include AI frameworks, parallel filesystems (such as Lustre), MPI I/O, and checkpointing for both serial and parallel applications.
Eligibility
This free tutorial is open to all those who are interested in learning about optimizing I/O in serial and parallel applications.
About the Instructor
Ritu Arora is a 2022 BSSw Fellow and a faculty member in the Computer Science departments at Wayne State University and the University of Texas at San Antonio (UTSA). She is also the founder of Venra Tech Inc., a company that provides solutions for advanced computing, data management, visualization, AI, and IT infrastructure development. Prior to joining Wayne State University, Ritu worked at UTSA as the Assistant Vice President of Research Computing, and before that, she served as a Research Scientist at the Texas Advanced Computing Center (TACC) at UT Austin. Ritu obtained her Ph.D. in Computer and Information Science from the University of Alabama at Birmingham in 2010. She is passionate about promoting the use of technology for creating societal impact and actively engages in causes aimed at creating inclusive policies and communities.
