- Видео 124
- Просмотров 4 595
OSG
Добавлен 19 окт 2021
The OSG provides common service and support for resource providers and scientific institutions using a distributed fabric of high throughput computational services. Researchers use OSG resources to solve scientific problems by breaking them down into a very large number of individual jobs that can run independently. The OSG does not own resources but provides software and services to users and resource providers alike to enable the opportunistic usage and sharing of resources.
Questions? Direct them to support@opensciencegrid.org. We're happy to help you get started on OSG, offer OSG support, or address any operational issues!
Follow us on Twitter: @opensciencegrid
Questions? Direct them to support@opensciencegrid.org. We're happy to help you get started on OSG, offer OSG support, or address any operational issues!
Follow us on Twitter: @opensciencegrid
OSG Training: Workflows with Pegasus
An introduction to the Pegasus Workflow Management System, which is a useful tool for researchers needing to execute a large number of jobs or complex workflows. Attendees will learn how to construct and manage workflows, capabilities like automatic data transfers, and higher level tooling to analyze the workflow performance.
Просмотров: 54
Видео
OSG Training: Moving your data with Pelican (and the OSDF)
Просмотров 485 месяцев назад
The OSPool runs hundreds of thousands of jobs everyday, and transferring data can sometimes require lots of power and effort. This training video covers how to compute and transfer large data using the OSDF, and how to use pelican to create data federations. Recorded June 2024.
HTC23 Final Closing Remarks
Просмотров 13Год назад
To close HTC23, Miron Livny gives some final closing remarks. Slides: agenda.hep.wisc.edu/event/2014/contributions/28451/ All talks, speakers, and materials from HTC23: agenda.hep.wisc.edu/event/2014/contributions/ All materials from past HTC Weeks and European Workshop Meetings: htcondor.org/past_condor_week...
HTC23 Generalizing HEPCloud Split Starter for Multiple Sites and Experiments
Просмотров 13Год назад
Bruno Coimbra and his team discuss generalizing HEPCloud Split Starter (HPSS). They explain what HPSS is, why it's needed, and implementation. Coimbra and his team then talk about running test jobs on Summit (OLCF). Templating, batch system scripts, VO repositories, and VO account.conf are also covered. Slides: agenda.hep.wisc.edu/event/2014/contributions/28450/ All talks, speakers, and materia...
HTC23 E4S: Extreme-scale Scientific Software Stack
Просмотров 20Год назад
Professor Sameer Shende discusses the Extreme-scale Scientific Software Stack (E4S) and the Exascale Computing Project. They then break down how to get started with E4S on OSG-HTC and aspects E4S includes. They delve into key considerations when using E4S and support. Slides: agenda.hep.wisc.edu/event/2014/contributions/28449/ All talks, speakers, and materials from HTC23: agenda.hep.wisc.edu/e...
HTC23 Scientific Workflows with Pegasus
Просмотров 28Год назад
Karan Vahi discusses scientific workflows with Pegasus. They explain what workflows are and challenges of them across domains. Vahi also breaks down key Pegasus concepts and Pegasus's part in the ACCESS support strategy. Slides: agenda.hep.wisc.edu/event/2014/contributions/28448/ All talks, speakers, and materials from HTC23: agenda.hep.wisc.edu/event/2014/contributions/ All materials from past...
HTC23 The Condor hatches in Africa
Просмотров 14Год назад
Marco van Zwetselaar explains the arrival of HTCondor in Africa. He explains how his team was able to get two hpcs up and running in Africa, as well as the benefits that come with that. Slides: agenda.hep.wisc.edu/event/2014/contributions/28447/ All talks, speakers, and materials from HTC23: agenda.hep.wisc.edu/event/2014/contributions/ All materials from past HTC Weeks and European Workshop Me...
HTC23 Transitioning Users to SciTokens and Getting them Closer to HTCondor with Jobsub_lite
Просмотров 8Год назад
Shreyas Bhat provides a brief background on the jobsub project and past issues. They then dive into an overview on jobsub_lite, tokens in the software, adoption of the software, and sucesses. They end with relaying the lessons they've learned during this transition to SciTokens. Slides: agenda.hep.wisc.edu/event/2014/contributions/28446/ All talks, speakers, and materials from HTC23: agenda.hep...
HTC23 Tracing Service for Tracing-Driven Glidein Optimizations
Просмотров 5Год назад
Namratha Urs and her team discuss GildeinWMS and the current state of Telemetry in GWMS. They also explain motivations and possible solutions, as well as how distributed tracing works. The team then breaks down the GWMS tracing service architecture. Slides: agenda.hep.wisc.edu/event/2014/contributions/28445/ All talks, speakers, and materials from HTC23: agenda.hep.wisc.edu/event/2014/contribut...
HTC23 Refactoring GlideinWMS Credentials
Просмотров 9Год назад
Bruno Coimbra and his team discuss refactoring GildeinWMS credentials. They go over what GildeinWMS is as well as the related factory and frontend. Traditional X509 Authentication, token authentication, credential refactoring, and generic credentials are also covered. Slides: agenda.hep.wisc.edu/event/2014/contributions/28444/ All talks, speakers, and materials from HTC23: agenda.hep.wisc.edu/e...
HTC23 HTCondor at BNL SDCC
Просмотров 6Год назад
Kavin Casella, Tom Smith, and Christopher Hollowell discuss the Scientific Data and Computing Center (SDCC) at Brookhaven National Laboratory (BNL). They provide a brief overview of the center and relay how HTC is incorporated into the center. Slides: agenda.hep.wisc.edu/event/2014/contributions/28442/ All talks, speakers, and materials from HTC23: agenda.hep.wisc.edu/event/2014/contributions/ ...
HTC23 HTCondor at Fermilab
Просмотров 14Год назад
Farrukh Khan discusses HTCondor at Fermilab. He explains what the Fermi National Accelerator Laboratory (Fermilab) is, as well as HTCondor's function in the lab. Khan then explains the CMS and Dune global pools, as well as different facilities in Fermilab. Slides: agenda.hep.wisc.edu/event/2014/contributions/28441/ All talks, speakers, and materials from HTC23: agenda.hep.wisc.edu/event/2014/co...
HTC23 Reaching new scales in the CMS Global pool
Просмотров 3Год назад
Marco Mascheroni and his team discuss the CMS SI team evaluating scalability and why they do so. They also present the 2023 scalability tests they did and explain what it means. They end off with reasons as to why they must keep pushing the SI for higher scales. Slides: agenda.hep.wisc.edu/event/2014/contributions/28440/ All talks, speakers, and materials from HTC23: agenda.hep.wisc.edu/event/2...
HTC23 Intel Microarchitecture: Nomen Numen
Просмотров 7Год назад
Greg Thain discusses Intel Microachitecture and the importance of well-named terms. He delves into the powers of clarity, metaphor, and generation when naming complex concepts. Thain then applies this to the OSPool as well as how Intel and RedHat help fix this problem. Slides: agenda.hep.wisc.edu/event/2014/contributions/28439/ All talks, speakers, and materials from HTC23: agenda.hep.wisc.edu/...
HTC23 Developing Machine Learning for Metadata Standardization with HTCondor and Snakemake
Просмотров 23Год назад
Yuriy Sverchkov, Mark Craven, and Colin Dewey discuss the development of machine learning for metadata standardization using HTCondor and Snakemake. They provde some useful context on metadata and the task of standardizing it. Development workflow is also covered, as well as the strengths and weaknesses of using Snakemake and HTCondor. Slides: agenda.hep.wisc.edu/event/2014/contributions/28432/...
HTC23 Running HTCondor cluster jobs at remote sites
Просмотров 13Год назад
HTC23 Running HTCondor cluster jobs at remote sites
HTC23 TaskVine: Workflow for Data Intensive and Serverless Applications
Просмотров 6Год назад
HTC23 TaskVine: Workflow for Data Intensive and Serverless Applications
HTC23 Automatic benchmarking of worker nodes to optimize resource provisioning
Просмотров 3Год назад
HTC23 Automatic benchmarking of worker nodes to optimize resource provisioning
HTC23 Highly available HTCondor pools in Google Cloud
Просмотров 3Год назад
HTC23 Highly available HTCondor pools in Google Cloud
HTC23 Orbiters and Clouds: Bringing Mars Reconnaissance Orbiter Imagery to the People
Просмотров 3Год назад
HTC23 Orbiters and Clouds: Bringing Mars Reconnaissance Orbiter Imagery to the People
HTC23 Rubin Observatory USDF-at-SLAC Update and HTCondor-at-NCSA 2023
Год назад
HTC23 Rubin Observatory USDF-at-SLAC Update and HTCondor-at-NCSA 2023
HTC23 Radio Interferometric Imaging using a cluster of GPUs on PATh
Просмотров 1Год назад
HTC23 Radio Interferometric Imaging using a cluster of GPUs on PATh
HTC23 Keynote: Unveiling the Universe through Gravitational Waves
Просмотров 21Год назад
HTC23 Keynote: Unveiling the Universe through Gravitational Waves
Thanks bro