BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Drupal//recurring_events_ical//2.0//EN
BEGIN:VEVENT
UID:02909876-fe38-4145-b76a-8d24c71bdbd8@support.access-ci.org
DTSTAMP:20221104T200641Z
DTSTART:20221109T190000Z
DTEND:20221109T210000Z
SUMMARY:Workflow Automation Tools for Many-Task Computing
DESCRIPTION:Historically, high-performance computing (HPC), has primarily i
 nvolved tightly coupled simulations executed in synchronous fashion across
  nodes. More recently, other paradigms within research computing have beco
 me common where the workload is not a single large coupled task, but inste
 ad very many modest tasks with little to no dependency on each other. Thes
 e could be data processing and analysis tasks, machine-learning experiment
 s, bioinformatics tasks, or parameter searches in a calculation.\n\nIn the
 se many-task, high-throughput computing (HTC) workloads, researchers often
  attempt some form of parallelization within their application language (e
 .g., Python, R, or MATLAB ®). This is tedious, fraught with difficulty fo
 r most users, lacking flexibility, and is a distraction from the research 
 concerns of the project. It would be better to use a distinct workflow aut
 omation tool to manage the many individual tasks.\n\nThis workshop outline
 s this paradigm on a traditional SLURM cluster along with several solution
 s. Different tools and techniques are discussed in an escalating fashion c
 overing various features and pitfalls. Finally, a demonstration of the hyp
 er-shell utility will showcase the level of sophistication possible both f
 or individuals and research groups in managing task execution. (hyper-shel
 l.readthedocs.io)
URL:https://support.access-ci.org/events/13
END:VEVENT
END:VCALENDAR