Agenda and Registration

Keynote: Monitoring and Analysis at Scale

Joost De Nijs, Facebook

Joost completed both his undergraduate and Master’s degrees from Brown University in Computer Science. While there he also worked in the VENLab (one of the first VR labs at the time) as their main developer. Since graduating, Joost has worked at Microsoft, Amazon and now FB focused mostly on large scale infrastructure (Azure, EC2, and now Data Infra at FB). His current role is as a Production Engineer on the Data Infra Foundations team at FB working on improving the reliability of the Batch workloads that power FB’s Big Data analytics.


Agenda (all times Pacific)

6:30am - 7:00 -- Meeting Access. Please join early to address any difficulties.

7:00am - 7:10 -- Opening (Brandt)

7:10 - 7:30 -- An Execution Fingerprint Dictionary for HPC Application Recognition. Jakobsche, Lachiche, Cavelan, Ciorba. Univ of Basel, Switzerland and Univ of Strasbourg, France (short paper)

7:30 - 8:00 -- An Integrated Job Monitor, Analyzer, and Predictor. Pal, Malakar. Indian Institute of Technology Kanpur, India (full paper)

8:00 - 9:00 -- Keynote: Monitoring and Analysis at Scale. De Nijs (Moderator: Lueninghoener)

9:00 - 9:10 -- Break

9:10 - 9:30 -- Backfilling HPC Jobs with a Multimodal-Aware Predictor. Lamar, Goponenko, Peterson, Allan, Brandt, Dechev. Univ of Central Florida and Sandia National Laboratories, United States (short paper)

9:30 - 10:00 -- Sequence-RTG: Efficient and Production-Ready Pattern Mining in System Log Messages. Harding, Wernli, Suter.IN2P3 Computing Centre, CNRS, Villeurbanne, France (full paper)

10:00 - 10:20 -- The Challenge of Disproportionate Importance of Temporal Features in Predicting HPC Power Consumption. Li, Karimi, Shin, Qi, Wang. Univ of Tennessee and Oak Ridge National Laboratories, United States (short paper)

10:20 - 10:40 -- Dynamic and Adaptive Monitoring and Analysis for Many-task Ensemble Computing. Jha, Malony. Rutgers Univ and Univ of Oregon, United States (short paper)

10:40 - 10:50 -- Break

10:50 - 12:00 -- Panel: Avoiding Misleading Analytic Results. Cory Lueninghoener (LANL) Moderator

  • Nicolas Lachiche, University of Strasbourg (France) - N. Lachiche is an associate professor in computer science and the head of the Data Science and Knowledge research group at the University of Strasbourg, in France. Since his PhD in 1997, his research has focused on mining complex data, typically relational data, on the methodology of machine learning, and on applications in various domains: chemistry, environment, health, and recently HPC monitoring.

  • Thomas Jakobsche, University of Basel (Switzerland) - Thomas Jakobsche is a PhD student at the University of Basel in the HPC group of the Department of Mathematics and Computer Science. His research interests include data analysis and machine learning methods and their use in HPC systems and applications. He works on performance optimization of parallel and distributed applications, as well as improving HPC operations and research through monitoring and operational data analytics.

  • Derek Tucker, Sandia National Laboratories (USA) - J. Derek Tucker is a Principal Member of the Technical Staff at Sandia National Laboratories. He received his B.S. in Electrical Engineering Cum Laude and M.S. in Electrical Engineering from Colorado State University in 2007 and 2009, respectively. In 2014 he received the Ph.D. degree in Statistics from Florida State University In Tallahassee, FL under the co-advisement of Dr. Anuj Srivastava and Dr. Wei Wu. He currently is leading research projects in the area of satellite image registration and statistical functional modeling of optical emissions. His research is focused on pattern theoretic approaches to problems in image analysis, computer vision, and signal processing.

  • Cari Martinez, Sandia National Laboratories (USA) - Cari Martinez is a senior computer scientist and the lead of the Applied Machine Intelligence research science team at Sandia National Laboratories. Her research focuses on developing deep learning methods to improve modeling capabilities and solve critical national security problems. She holds a BS in Honors Mathematics from the University of Notre Dame, an MS in Computer Science from the University of New Mexico, and she is currently finishing her Computer Science PhD at Arizona State University under the advisement of Dr. Stephanie Forrest.

Registration:

All presenters and attendees must register at the IEEE Cluster website.

Zoom Link:

  1. When you register, you will receive an email from IEECS <noreply@rdmobile.com> with a link to "Access Event on Desktop"

  2. Click on that link and you will be in the Cluster RDMobile Platform

  3. Either Click on "Schedule" and Search for HPCMASPA OR Click on "Schedule" -> "Workshop"->"HPCMASPA". This will get to you to the zoom link.

  4. Also Click on "Slack Workspace" and join "Cluster 2021" and the "workshop-hpcmaspa" channel.