Cloud Software Development Engineer
Intel Corporation
Wrote an advanced
AI
/
GPU
profiler for Intel GPUs, and
open-sourced it. It uses eBPF,
kernel driver interfaces, and new hardware sampling features to
present flamegraphs and flamescopes of kernels running on the GPU.
Wrote a per-process instruction mix profiler, and
open-sourced it.
It uses perf events and minimal eBPF to sample instructions being
executed on the system, disassemble them, and present them to the
user in an easy-to-understand format.
-
Performed performance analysis, benchmarking, and optimization work for large cloud customers
(Netflix, Pinterest, Disney Streaming).
|
August, 2021 - Present
|
Graduate Research Assistant
The University of Tennessee
Developed a compiler pass and runtime library to enable efficient
utilization on heterogeneous memory technologies.
Modified the Linux mm subsystem and the HotSpot Java virtual machine
to control the placement of hot and cold data and reduce energy consumption
on systems with multiple DRAM ranks.
|
June, 2015 - August, 2021
|
Graduate Research Intern
Intel Corporation
|
July, 2019 - July, 2020
|
ASTRO Graduate Intern
Oak Ridge National Laboratory, CSR Group
|
July, 2018 - February, 2019
|
ASTRO Graduate Intern
Oak Ridge National Laboratory, CSR Group
Developed novel offline and online profiling and analysis strategies for
understanding application data usage.
Created runtime extensions and a custom memory allocator to incorporate memory
usage profiles during heterogeneous memory management.
|
January, 2018 - May, 2018
|