Digistor accelerates the Australian Government’s Research Data Storage Infrastructure Project with Aspera
The Australian Government Initiative, the Research Data Storage Infrastructure (RDSI) Project, working closely with the Australian Government's National Research Network (NRN) Project, is deploying solutions from Aspera to move large data to, from and between nationally-dispersed storage nodes and high-performance computing centers to strengthen Australia's capabilities in data-intensive research collaboration.
Funded through the Education Investment Fund under the Super Science (Future Industries) initiative, and led by the University of Queensland, the RDSI project has received A$50 million in federal funding to provide data storage and collaboration infrastructure for Australia's research community. The project is intended to assist institutions and researchers to more easily and effectively access, use, manage and share large holdings of research data, expanding the scale and scope of problems that Australian researchers may seek to address.
The network consists of eight high-performance data storage locations throughout Australia, with associated Data Transfer Nodes - clusters of high-performance host computers and storage - connected by a fast network optimized for high-performance scientific applications. Aspera Connect Server software is being deployed on all hosts in each Data Transfer Node in a high availability configuration to provide ultra high-speed transfers regardless of data size and network distance.
The Aspera Shares web application was selected to provide a browser-based, user-friendly overlay for uploading and downloading file content to and from the Nodes and fine-grained control for researchers to manage access to their data, with built-in authentication for the LDAP and single sign-on SAML services, common throughout the research community.
For initial data ingest, researchers simply upload data to the Aspera Connect Server clusters at each Node using the freely downloadable Connect Web Browser Plug-in which integrates directly with all browser platforms to support high-speed fasp transfers of any file size directly in the browser or from the command line. Once ingested, Aspera Shares provides easy access to shared data with a single web interface to consolidate browsing of shared content across all nodes, enabling fast collaboration between researchers at major universities throughout Australia.
In addition to speed and security, the Aspera technology provides complete control over transfer bandwidth and finish times, and is robust to network outages with automatic retry and resume of even the largest data sets.
During the pilot deployment of the Aspera solution, a research group at a university located in central Queensland, with relatively limited network bandwidth, had an urgent requirement to transfer approximately 1.1TB of data overnight to a government agency in the state capital of Brisbane, where one of the RDSI Nodes is located. The bandwidth management and automatic resume features of the Aspera solution were critical in meeting the deadline, as the transfer could use only 40% of the available bandwidth, and the network link was shut down for several hours during the night for preventative maintenance.
RDSI also selected Aspera Console, the web-based management application, to provide complete visibility over the Aspera transfer environment, including real-time, centralized control over transfers, nodes and users.
"To complement open source data transfer options, we were looking for a commercial solution that is widely available, meets our stringent performance requirements, and is very easy to use. So we deployed Aspera as the data mover for our nodes, and layered the Shares web interface over the top so researchers could more easily access and share their big data. Large file transfer speeds over the fastest parts of the infrastructure are full line speed (10 Gigabits per second) and international transfers with the US have benchmarked at 7-8 Gigabits per second. In addition, the management platform provides us complete visibility into transfer performance," said Peter Elford, RDSI network program manager.
Another component of the RDSI initiative is to explore the use of cloud storage to support data-intensive research and the researchers who must use, manage and share large and ever-growing data sets. Aspera's support of cloud-based object storage and commercial Infrastructure as a Service (IaaS) platforms will be explored to determine if and what role cloud services can play in supporting the infrastructure already deployed at the existing storage locations.
"In the research sector, data is everything. Researchers need to have complete and immediate access to the data to be successful and the data sizes are exploding. Aspera software enables scientific researchers to access shared data around the globe while maintaining maximum security and performance," said Michelle Munson, CEO and co-founder of Aspera. "We are proud and very excited to be a part of the RDSI project and look forward to bringing high-performance transfer software to support its ambitious initiatives for the Australian research community."
“RDSI is moving large data sets around the country faster than it can be copied off a USB hard drive to a local desktop,” said Matt Wood, Account Manager at Digistor. “They have been very successful in eliminating bottlenecks in their infrastructure and maximising through put using Aspera technology”.
The Research Data Storage Infrastructure (RDSI) Project, an initiative of the Department of Education, is funded from the Education Investment Fund under the Super Science (Future Industries) initiative.
RDSI is a $50m Australian Government funded project, for which the University of Queensland (UQ) is the lead agent.