Durga Keerthi Mandarapu

Durga Keerthi Mandarapu

Ph.D. Candidate Computer Science

Purdue University

Hey!

I’m looking for full-time research positions. Please reach out to me, if you think my expertise fits the role you are hiring for.

I am a PhD student in Computer Science at Purdue Univeristy, advised by Prof. Milind Kulkarni . My thesis introduces several reductions to accelerate applications involving irregular programs by mapping them to the hardware acceleration provided by GPU Ray-Tracing architecture. My broad research interests span Parallel Computing, High-Performance Computing, Databases, and Compilers. Before coming to Purdue, I did my undergrad in Computer Science & Engineering at the Indian Institute of Technology, Hyderabad, India.

I recently finished my Summer ‘24 Internship at Meta, where I worked on building a scalable distributed random walks service on a large scale graph. Prior to this, in the summer 2022, I interned at Katana Graph, where I had the opportunity to work with Roshan Dathathri on accelerating random walks on graphs. I was also at Simon Fraser Univerity, Vancouver, Canada briefly during the summer 2019 for MITACS internship. Thanks to Prof. Keval Vora, I had an amazing experience building a scalable solution to process computations on streaming graphs.

Interests
  • Parallel Computing
  • High-Performance Computing
  • Databases
  • Compilers
Education
  • Ph.D. in Computer Science, Spring 2025*

    Purdue University

  • B.Tech(Honors) in Computer Science and Engineering, Minor in Economics, Spring 2019

    Indian Institute of Technology, Hyderabad

Publications

(2024). S-ray: Accelerating Spatial Queries using GPU Ray Tracing.

Cite

(2024). Arkade: k-Nearest Neighbor Search With Non-Euclidean Distances using GPU Ray Tracing. International Conference on Supercomputing [Best Paper Award].

PDF Cite Code DOI

(2024). Mochi: Fast & Exact Collision Detection.

PDF Cite

(2023). Accelerating Unbounded k-Nearest Neighbor Search using RT Cores. International Conference on Supercomputing.

PDF Cite Code DOI

(2017). EpiStrat: A Tool for Comparing Strategies for Tackling Urban Epidemic Outbreaks. International Conference on Smart Health.

PDF Cite Code

Internships

 
 
 
 
 
Meta
Software Engineering Intern
May 2024 – August 2024 Bellevue
To build scalable distributed random walks applications, I separated the compute and storage tiers and benchmarked the new service along several performance metrics.
 
 
 
 
 
Katana Graph
Software Engineering Intern
May 2022 – September 2022 Remote
I worked on developing a scalable uniform random walks application to overlap communication and computation costs on distributed graphs using the Katana interface.
 
 
 
 
 
Simon Fraser University
MITACS Intern
May 2019 – August 2019 Vancouver, Canada
I developed a parallel incremental algorithm that processes non-monotonous dynamic edge updates to compute a betweenness centrality measure of all the vertices in a streaming graph.
 
 
 
 
 
Purdue University
Purdue Undergraduate Research Intern
May 2019 – August 2019 West Lafayette, IN
I developed a credit network using smart contracts in Ethereum that allow payments across different currencies without introducing a new crypto-currency and at a lowered account-creation, direct-payment, and currency transaction costs.

Service

  • Reviewer, Transactions on Big Data 2024
  • Mentor, SIGPLAN long‑term mentorship Program, PLDI 2024
  • Student Volunteer, PLDI 2023
  • Student Volunteer, SOSC 2019
  • EML Web Coordinator, IIT Hyderabad, 2017-18
  • Vidyadhaan Coordinator, National Servie Scheme, 2017-18
  • Web Coordinator, IIT Hyderabad Alumni Cell, 2016-17

Teaching

  • Graduate Teaching Assistant, Data Structures, Purdue University (Summer 2021, Spring 2021, Fall 2020, Spring 2020, Fall 2019)
  • Teaching Assistant, Operating Systems, IIT Hyderabad (Spring 2019, Fall 2018)
  • Teaching Assistant, Database Systems, IIT Hyderabad (Spring 2019)
  • Teaching Assistant, Data Structures, IIT Hyderabad (Fall 2017)
  • Teaching Assistant, Introduction to Programming, IIT Hyderabad (Fall 2017)

Contact