On this Saturday, September 18th, 2021, our weekly scientific seminar of the HPC Lab Students will be held. This seminar will continue examining Big Data tools and algorithms, presented by both graduated and undergraduate students. After this seminar, students will be able to apply this knowledge to analyze data in their thesis and further research in the HPC Lab.

I. Speakers and Contents:

1. Apache Spark Programming Guide 

Abstract:

Apache Spark is a unified analytics engine for big data processing, with built-in modules for streaming, SQL, machine learning, and graph processing. In this talk, we’ll focus on the best practice in Spark and discuss how to contribute new algorithms to Spark.

2. Deep Reinforcement Learning for HPC Schedulers

Abstract:

Efficient job scheduling is crucial to optimize productivity in high-performance computing. Existing HPC often uses heuristics to estimate and schedule jobs. But this approach lacks scalability as it can’t be modified to adapt to new changes in the system. Moreover, job workloads may shift month by month and the optimization goals may also vary across time. RLSheduler, an automated HPC batch job scheduler built on reinforcement learning, can learn on its own without manual interventions or expert knowledge.

II. Event Information and Registration:

The language of this seminar is Vietnamese. Due to the outbreak of the COVID-19, the seminar will still be held online on the Google Meets platform. Registered participants will receive an email with an invitation to the online meeting.

Registration for the seminar is available here at no cost. However, due to the limitation of Google Meets (only 100 people per meet), available slots are limited.

However, HPC Lab will provide you the recorded links after the seminar if you are unable to attend this event. For more information, please contact hpcc@hcmut.edu.vn.

We are looking forward to seeing you! 

The HPC Lab Team.

This event is organized by HPC Lab (HCMUT, VNU HCM) – The Leading HPC Research Team in Viet Nam.