Parianz Barakhshan

 

Parianz Barakhshan

Ph.D. Candidate and Research Assistance @University of Delaware, USA

Department of Electrical and Computer Engineering

Parinazb [at] udel [dot] edu

Linkedin

Member of:

Conferences Attended:

Posters:

Student Volunteer:

  • PEARC19

 

parinazb[at]udel[dot]edu

Linkedin

GitHub

Website

I am a second-year Ph.D. candidate and Research Assistant at the University of Delaware, Department of Electrical and Computer Engineering.

My research interests include:

  • optimizing compilers,
  • programming methodologies,
  • performance evaluation for high-performance computing,
  • Best Practices & tools in CDI(Computational and Data-intensive) research,

I have always loved learning, growing, developing, and challenging myself so that I can contribute to something significant. That is why you can see through my resume working for the industry for a while and then going back to Academia.
At my current role in our research group, we look into the requirements that our customers or other research groups bring in, and we help them with developing, optimizing, and running their Computational and Data-intensive(CDI) applications. Let me summarize it all by saying that my purpose is to help the world run better.

Currently, I am looking for internship opportunities for this summer. I am interested in working on parallelizing computational and Data-intensive (CDI) applications and learning more about parallelization techniques ad implementing those techniques on source-to-source parallelizing compilers Infrastructures.

I am involved in different projects that I explain below:

The Xpert NetworkThe community you want to join to get the support you need in developing, optimizing, and running CDI applications.

I am currently leveraging a project called the Xpert Network that has already put me in contact with many national activities in HPC application development. The Xpert Network project is a national level effort providing the support infrastructure required by all those involved in Computational and Data-intensive research.

You can be a

  • domain Scientist or
  • an Xpert who gives direct assistance to CDI researchers or
  • a tool developer that facilitate CDI application development.

This is the platform to talk to your peers about:

  • the best practices you are applying in your research team,
  • the tools you are using,
  • the challenges you are facing, or
  • your open problems.

And get help or the feedback of those in the field.

Lots of activities are happening in this network that can be helpful and informative to people involved in the CDI research. 

Project Status: Active

Online portal for high-precision atomic physics data and computation

This is a collaborative project with the physics department of UD. The goal of this project is to create an online portal for high-precision atomic physics data and computation that will provide a variety of services to address the needs of the widest possible community of users. The first version of the portal, demonstrated here, provides a wide range of transition matrix elements, static, and dynamic polarizabilities for a number of atoms and ions. The data are calculated using a high-precision state-of-the-art coupled-cluster (CC) method or a hybrid method that combines configuration interaction and CC. We seek community input to improve the portal and guide the next stages of the project which will include more complicated systems and capabilities to compute atomic properties on-demand via the portal.

Project Status: This project is not publicly available yet.

AutoTuner for Cetus Compiler

In this project, we created an automatic tuning program written in a bash script that utilizes the Cetus source-to-source compiler to find the fastest combination of optimization options for NAS Parallel BE Apps to optimize in parallel.

In this project, we created an automatic tuning program written in a bash script that utilizes the Cetus source-to-source compiler for running a different combination of optimization options on top of  NAS Parallel Benchmark Applications. We have utilized the UD Caviness Cluster for compiling and running this Computational and Data-intensive program. The goal of this project is to check the dependency of different parameters and their effect on different Benchmark applications.

Benchmark: NAS Parallel Benchmarks

Cetus: A Source-to-Source Compiler Infrastructure for C Programs

Project Status: Still in process of Producing and analyzing data.

 

Interactive Parallelizing Compiler

Today’s computers are all Multicores. With parallelization techniques, we can convert sequential code into multi-threaded and/or vectorized code in order to use the power of multiple processors simultaneously in a modern shared-memory Architecture.

This tool that we are developing allows users to apply parallelization techniques in a semi-automatic way. While the tool can parallelize the code/ parts of the code in a fully automatic format for Non-expert users/end users,  Experts/Power-users can steer the parallelization process through the tool menus to get what they want from the tool.

The underneath compiler infrastructure used in this project is Cetus. Cetus provides the source-to-source compiler infrastructure for C programs, and it was previously developed in my research group. What Cetus does is transforming the code so that a program can be executed in parallel.

Cetus does it all through executing a set of analyses. It identifies the part of the program that can be parallelized, adds the related OpenMP directives to those program sections. Then it transforms the code in an automatic way so that a program can be executed in parallel.

An initial prototype of an interactive parallelizer is part of the current Cetus release.

Cetus: A Source-to-Source Compiler Infrastructure for C Programs

Project Status: In progress.

I am interested in software development for high performance and distributed computing systems. My purpose is to help the world run better.