about-core

Open Science Data Cloud Automate Provisioning

OSDC (Open Source Data Cloud) provides tools so that researchers can upload and analyze datasets, integrate their data with data from research collaborators, and integrate their data with public data for analysis. Members include universities, (University of Chicago, Florida International University, University of Illinois Chicago, Northwestern University), companies, (Cisco and Yahoo), government agencies (NSF, NASA) and foreign partners (University of Edinburgh, UNESP, UFF, BIG, University of Amsterdam, AIST, etc.). The OSDC currently hosts more than 600 TB of public datasets from a number of disciplines, published online and available to users.

The OSDC user services include the ability to provision virtual machines, access usage and billing information, share files, and access to a key service and public datasets. All of the OSDC user services are tied together through a customizable web application “Tukey” and middleware that enables uniform access to the cloud services by the “Tukey Console” web application. The OSDC is a distributed facility connected by 10G or greater networks so high speed transport protocols are important for enabling users to import/export data and to move data around flexibly in their analysis processes.

Full rack installation of “OpenStack” manually takes over a week to complete. Whereas the OSDC provides fully automate provisioning, taking a rack from bare metal to a compute or storage cloud in much less than a day.

Speakers

Authors

  • Robert Grossman, Heidi Alvarez, Paula Grosso, Adam Barker,
  • Vasilka Chergarova

Part of session

Lightning talks

Related documents