HTCondor Week 2017

University of Wisconsin — Madison, Wisconsin — May 2-5, 2017

picture of Madison

Thursday, May 4th

 8:00 am  9:00 am Coffee and Registration
coffee, tea, ice water
Session Moderators: Brian Lin, Mark Coatsworth
 9:00 am  9:20 am The Dynamic Scheduling Strategy of HTCondor at IHEP
Jingyan Shi
Institute of High Energy Physics, Chinese Academy of Science
 9:25 am  9:45 am Disk-to-Disk and Day-to-Day Data Placement Performance Metrics on a Trans-pacific HTCondor Infrastructure
Philip Papadopoulos
University of California-San Diego
 9:50 am 10:10 am Case Study: UW-Madison CHTC Scheduling Policies
Greg Thain
Center for High Throughput Computing
10:10 am 10:40 am Break
Session Moderators: Kent Wenger, Mark Coatsworth
10:40 am 11:00 am Detecting and Managing Job Progress and Events with an HTCondor Update Job Info Hook
Michael Pelletier
Raytheon Company
11:05 am 11:25 am Monitoring HTCondor
Todd Tannenbaum
Center for High Throughput Computing
11:30 am 11:50 am Monitoring and Analytics with HTCondor Data
William Strecker-Kellogg
11:55 am 12:15 pm Monitoring HTCondor using Elastic, Grafana, and Conmon
William Deck
Susquehanna International Group
12:15 pm  1:15 pm Lunch
Session Moderators: Carl Edquist, Matyas Selmeci
 1:15 pm  1:35 pm Google Compute Engine with HTCondor
Karan Bhatia
 1:40 pm  2:00 pm Using Docker, HTCondor, and AWS for EDA Model Development
Andy Howard
Cycle Computing
 2:05 pm  2:25 pm Technical Questions Raised During Project “Interfacing HTCondor-CE with OpenStack”
Jose Caballero
 2:30 pm  2:50 pm High-Throughput Machine Learning from Electronic Health Records
David Page
 2:55 pm  3:15 pm HTCondor in the Azure Cloud
Taylor Newill, Jaime Frey
Tata Institute of Fundamental Research, Microsoft, Center for High Throughput Computing
 3:15 pm  3:45 pm Break
Session Moderators: Jason Patton, Erin Grasmick
 3:45 pm  4:05 pm Applications of HTCondor to the Environmental Sciences
Environmental scientists face computational hurdles. Large models must be run and parameterized. -omics data must be processed. Landscape-level studies must be designed and analyzed. High-throughput computing, and more specifically HTCondor, is one approach that may be used to solve these problems. In this talk, I describe how HTCondor is being used by scientists at USGS to solve these problems.
Richard Erickson
USGS/Upper Midwest Environmental Sciences Center
 4:10 pm  4:30 pm Pegasus WMS : An Introduction and Upcoming Features
Karan Vahi
USC Information Sciences Institute
 4:35 pm  5:05 pm Automation Panel
Miron Livny, others
Center for High Throughput Computing
 5:05 pm  5:10 pm Closing Remarks
Miron Livny
Center for High Throughput Computing

Specific talks and times are subject to change.

HTCondor Week Social Gathering at the Memorial Union Terrace

We will have an informal gathering following the Thursday session. It will be held at the Memorial Union Terrace on the UW-Madison campus (the Memorial Union Rathskeller in case of bad weather), a short walk from the Fluno Center. More details to come!