CV

Education

• Doctoral Student in Computer Science, Texas Tech University, 2023 (expected)
• M.S. in Computer Science, Texas Tech University, 2019

Work experience

• Fall 2019 - Present: Graduate Student Research Assistant
  • DISCL at Texas Tech University
  • Supervisor: Dr. Yong Chen
• Summer 2022: Graduate Student Intern
  • National Energy Research Scientific Computing Center (NERSC)
  • Duties included:
    • Analysis of Permutter monitoring metrics.
  • Supervisor: Dr. Brandon Cook, Dr. George Michelogiannakis
• Summer 2021: Graduate Student Intern
  • National Energy Research Scientific Computing Center (NERSC)
  • Duties included:
    • Analyzed the job-level monitoring data and applied Machine Learning models (Random Forests, SVC, LinearSVC, etc.) to classify the jobs using statistical features extracted from the time-series data.
    • Explored the method of encoding the time-series monitoring data into images (named job signatures); trained a Convolutional Neural Networks to classify the job signatures.
  • Supervisor: Dr. Brandon Cook

Skills

• Programming Languages
  • C/C++, Bash, Python
• Data Analysis Libraries
  • Numpy
  • Pandas
  • Scikitlearn
  • Matplotlib
• Databases
  • PostgreSQL
  • TimescaleDB
  • InfluxDB
• Operating systems
  • UNIX/Linux
  • Mac OS

Publications

MAC: Memory access coalescer for 3D-stacked memory

Wang, Xi, Antonino Tumeo, John D. Leidel, Jie Li, and Yong Chen. "MAC: Memory access coalescer for 3D-stacked memory." In Proceedings of the 48th International Conference on Parallel Processing, pp. 1-10. 2019.

PIMS: a lightweight processing-in-memory accelerator for stencil computations

Li, Jie, Xi Wang, Antonino Tumeo, Brody Williams, John D. Leidel, and Yong Chen. "Pims: a lightweight processing-in-memory accelerator for stencil computations." In Proceedings of the International Symposium on Memory Systems, pp. 41-52. 2019.

Mtsad: Multivariate time series abnormality detection and visualization

Pham, Vung, Ngan Nguyen, Jie Li, Jon Hass, Yong Chen, and Tommy Dang. "Mtsad: Multivariate time series abnormality detection and visualization." In 2019 IEEE International Conference on Big Data (Big Data), pp. 3267-3276. IEEE, 2019.

RadarViewer: Visualizing the dynamics of multivariate data

Nguyen, Ngan, Jon Hass, Yong Chen, Jie Li, Alan Sill, and Tommy Dang. "RadarViewer: Visualizing the dynamics of multivariate data." In Practice and Experience in Advanced Research Computing, pp. 555-556. 2020.

MonSTer: an out-of-the-box monitoring tool for high performance computing systems

Li, Jie, Ghazanfar Ali, Ngan V. T. Nguyen, Jon Hass, Alan Sill, Tommy Dang and Yong Chen. “MonSTer: An Out-of-the-Box Monitoring Tool for High Performance Computing Systems.” 2020 IEEE International Conference on Cluster Computing (CLUSTER) (2020): 119-129.

HAM: Hotspot-Aware Manager for Improving Communications with 3D-Stacked Memory

Wang, Xi, Antonino Tumeo, John D. Leidel, Jie Li, and Yong Chen. "HAM: Hotspot-Aware Manager for Improving Communications with 3D-Stacked Memory." IEEE Transactions on Computers 70, no. 6 (2021): 833-848.

Posters

• HPC Application Recognition Through Image-encoded Monitoring Data (Best Poster Award at Spring 2022 NSF Cloud and Autonomic Computing Industry Advisory Board Conference)
• Detecting and Identifying Applications by Job Signatures in HPC Systems (Research Poster, accepted by SC21)

Paper Review

• 2021 IEEE International Conference on Big Data
• 2020 IEEE International Conference on Big Data

Talks

Monitoring Power Usage of Jobs Running on Quanah Cluster

September 18, 2019

Seminar at DISCL, Lubbock, Texas

PIMS-A Lightweight Processing-In-Memory Accelerator for Stencil Computations

October 01, 2019

Conference at The International Symposium on Memory Systems, Washington, D.C.

Profiling Power Consumption of Jobs with SLURM

January 29, 2020

Seminar at DISCL, Lubbock, Texas

Collecting and Storing Telemetry Metrics from RedRaider Cluster

February 27, 2020

Seminar at DISCL, Lubbock, Texas

Experiences of Storing and Querying Monitoring Data of Large-scale High Performance Computing Platforms

April 15, 2020

Seminar at DISCL, Lubbock, Texas

Monitoring Operating System Status (on a Raspberry Pi cluster)

July 22, 2020

Seminar at DISCL, Lubbock, Texas

MonSTer: An Out-of-the-Box Monitoring Tool for High Performance Computing Systems

September 16, 2020

Conference at The 22nd IEEE International Conference on Cluster Computing, Kobe, Japan

The IEEE Cluster2020 Experience, MonSTer Review and Future Work

September 23, 2020

Seminar at DISCL, Lubbock, Texas

Predicting Abnormal Workloads in HPC Systems

May 08, 2021

Seminar at DISCL, Lubbock, Texas

Detecting and Identifying Applications by Job Signatures

September 14, 2021

Seminar at DISCL, Lubbock, Texas

Advanced Visualization and Data Analysis of HPC Cluster and User Application Behavior

November 12, 2021

Conference at The International Conference for High Performance Computing, Networking, Storage, and Analysis, Saint Louis, Missouri

Deep Learning and Monitoring Metrics to Image Encoding for Detecting Applications in HPC systems

November 23, 2021

Seminar at DISCL, Lubbock, Texas

Lightweight Checkpointing of Loop-Based Kernels Using Disaggregated Memory

February 02, 2022

Seminar at DISCL, Lubbock, Texas

SST and Cycle-accurate Simulation of xBGAS

May 03, 2022

Seminar at DISCL, Lubbock, Texas

A Holistic View of Resource Utilization on Perlmutter

August 03, 2022

Seminar at DISCL, Lubbock, Texas

Research Experience

• Collecting and Storing Telemetry Metrics from HPC cluster
  • Investigated the Integrated Dell Remote Access Controller (iDRAC) and explored pull model and push model (subscription) for obtaining remote monitoring data
  • Developed a set of tools for configuring iDRAC9 telemetry reports and initializing tables in TimescaleDB
  • Implemented the code to asynchronously collect telemetry metrics from 240 CPU nodes and 20 GPU nodes via Server-Sent Event (SSE), collected monitoring metrics from 467 CPU nodes via pull model, and utilized the resource manager (e.g. Slurm, UGE) interface to collect job accounting information
• Improving the Monitoring Data Querying Performance
  • Optimized the database schemas in InfluxDB, reducing 71.98% of data volume and gaining up to 1.76X performance boost
  • Developed a middleware between the consumers (i.e. visualization and automation components) and the producers (i.e. databases) called Metrics Builder
  • Metrics Builder hides the details of querying databases, speed up querying and transmitting, and provide a unified API (based on OpenAPI) and dataframe for data analysis consumers
  • Metrics Builder improves query performance by up to 25X and reduces the volume of data transmitted by 95%
• Processing-in-Memory Accelerator for Stencil Computations
  • Investigated the cache performance of stencil computation; implemented a simulator generating the memory traces of different types of stencils
  • Designed a Processing-In-Memory (PIM) accelerator based on the Hybrid Memory Cube (HMC) architecture
  • Evaluated the proposed accelerator using HMC-Sim library. The results showed that the proposed accelerator reduces 48.25% of data movement on average and obtains up to 65.55% of bank conflict reduction
• Memory Power Tracing on Hybrid Memory Cube (HMC)
  • Investigated the impacts of the memory accesses on the power consumption in Hybrid Memory Cube (HMC)
  • Built a Power-Sim driver utilizing the HMC-Sim library to simulate the memory power consumption, run the driver under the risc-v spike simulator
  • Evaluated the efficacy of the Dynamic Memory Coalescer(DMC) based on the simulation results

Awards

• Summer Thesis/Dissertation Research Award (Mar. 2019)