2018 Technical Program
8:45-9:00 AM - Workshop Opening
-
Workshop Opening
Walfredo Cirne, Narayan Desai and Dalibor Klusáček
9:00-10:00 AM - Keynote Lecture
- Google Flex.
John Wilkes (Google, Inc.)
Abstract: Ensuring that users in a computation cluster can receive strong guarantees about immediate resource availability is surprisingly hard to do without wasting significant amounts of resources. The Google Flex system gives administrators a way to define resource pools that provide a range of strong, statistically-backed guarantees based on user and job behavior, while also reducing human effort by automation tied into the Borg cluster manager, Colossus distributed file system, and many other services. Flex has been widely adopted at Google, and has produced significant resource savings.
Biography: John Wilkes has been at Google since 2008, where he is working on automation for building warehouse scale computers. Before that, he spent a long time at HP Labs, becoming an HP and ACM Fellow in 2002. He is interested in far too many aspects of distributed systems, but a recurring theme has been technologies that allow systems to manage themselves.
10:00-10:30 AM - Coffee Break
10:30 - 12:00 AM - Session 1
- Analysis of Job Metadata for Enhanced Walltime Prediction.
Mehmet Soysal, Marco Berghoff and Achim Streit - Evaluating the Impact of Soft Walltimes on Job Scheduling Performance.
Dalibor Klusáček and Václav Chlumsky (CESNET) - Reducing the Human-in-the-Loop Component of the Scheduling of Large HTC Workloads.
Frederic Azevedo and Frederic Suter
12:00-1:30 PM - Lunch Break
1:30 - 3:00 PM - Session 2
- Using Pilot Systems to Execute Many Task Workloads on Supercomputers.
Shantenu Jha, Matteo Turilli, Andre Merzky and Manuel Maldonado - Adaptive Simultaneous Multi-tenancy for GPUs.
Ramin Bashizade, Yuxuan Li and Alvin Lebeck - Stochastic Programming Approach for Resource Selection under Demand Uncertainty.
Tanveer Bhuiyan, Mahantesh Halappanavar, Ryan Friese, Hugh Medal, Luis de La Torre, Arun Sathanur and Nathan Tallent
3:00-3:30 PM - Coffee Break
3:30 - 5:00 PM - Session 3 + Panel
- Approaching Actor-Level Resource Control for Akka.
Ahmed Abdelmoamen, Dezhong Wang and Nadeem Jamali - Panel discussion
JSSPP attendees
JSSPP 1995-2024 – the Workshop on Jobs Scheduling Strategies for Parallel Processing. Contact email: jssppw@gmail.com