A key goal in the Open Science Grid (OSG) is to increase the access to distributed high-throughput computing (DHTC) for researchers at U.S. campuses. OSG is a powerful sharing ecosystem that enables research communities to access and share computing resources in flexible ways, but there can be a steep learning curve for those not already familiar with DHTC. To lower the entry barriers, we have built an easier on-ramp to the OSG that shields the user from the complexities of the grid. Today we have a set of entry methods that significantly reduce the effort needed to use OSG DHTC; these methods are depicted below.
Image Credit: Chander Sehgal
We know that one size does not fit all, and thus we have developed a spectrum of methods. For individual users we provide three methods: 1) users can contact OSG directly and get access via the OSG-Connect service that provides easy-to-use login access based on their existing campus credentials; 2) other users access OSG by applying thru XSEDE; and 3) users with advanced needs are enabled via the OSG-Direct login host. And we also support research groups and small communities who want to administer their own login nodes for their users by providing a “bulk” connection; an example is the Baker Lab at the University of Washington.
OSG members connect their computing resources both to perform computing for their own experiments and to make available what they do not use at that time to other OSG members; these shared cycles are referred to as “opportunistic cycles” and are the fuel that powers the OSG Open Facility. In the last year these “opportunistic cycles” have provided access to DHTC for about 75 researchers who used about 100 million CPU hours. This on-ramp to the OSG continues to grow in capability and use, and we are actively working to identify and connect more researchers to OSG DHTC.
~ Chander Sehgal