OSG is a consortium dedicated to the advancement of open science via the practice of distributed High Throughput Computing, and the advancement of its state of the art.
The OSG Consortium builds and operates a set of pools of shared computing and data capacity for distributed high-throughput computing (dHTC). Each pool is organized and operated to serve a particular research community (e.g. a campus, multi-institutional collaboration, etc.), using technologies and services provided by the core OSG Team. One of these pools, known as the Open Science Pool is operated for all of US-associated open science. The Consortium, thus, represents the totality of all researchers, resources, individuals and institutions that benefit from or contribute to any of the OSG Fabric of Services (further below).
The OSG Council governs the consortium ensuring that the OSG benefits the scientific mission of its stakeholders: the research communities, organizations that provide resources and services for them, including funding resources.
The Executive Team manages the core OSG Team, with team members from various institutions organized into Areas that provide core OSG technologies and services required to operate pools in support of research communities. As of January 2022, the OSG Team is funded primarily via the PATh (NSF #2030508), IRIS-HEP (NSF #1836650) projects, and in kind contributions from multiple entities, including DOE national laboratories.
Distributed High Throughput Computing
High-throughput computing (HTC) is the execution of computational work in the form of numerous, self-contained tasks to optimize their overall completion across available computing resources. Specialized by the OSG, distributed high-throughput computing (dHTC) involves the operation of HTC-optimized infrastructure across many independent, collaborating administrative domains.
The OSG Fabric of Services
The OSG provides various open-source software, other technologies, and services for researchers and research organizations to support their dHTC compute requirements. Our software and technologies allow research organizations to build dHTC systems at-scale from shared computing and data resources, and to make these resources available to researchers within virtual clusters (what the OSG calls “pools”).
Some pools or participating organizations provide application-specific interfaces or datasets, such that end-users may not even be aware of integration with OSG and its services/technologies. Many OSG technologies and services are based upon or directly leverage the HTCondor Software Suite (HTCSS), in addition to other open-source tools that enable shared computing and data capabilities. These can be deployed by a research organization to create their own dHTC pool across participating compute and data components, making their federated capacity available to the researchers they serve.
OSG Technologies and Services
The OSG’s Global Research Accounting (GRACC) system provides metrics of capacity contributions and usage across reporting OSG pools, including specific metrics for participating institutions and research projects, some of which represent hundreds or thousands of individual users.
The Open Science Pool (OSPool) provides dHTC capacity for research projects associated with a US-based academic, government, or non-profit organization, and with funding from the National Science Foundation through the PATh project (NSF #2030508). Applicable researchers can obtain access to OSPool resources via an access point operated by their institution or collaboration, or via the OSG Portal.
Learn more about other OSG Pools, who contributes, and who uses them.
Specific OSG Softwares are described in the OSG Site Admin documentation.