November 14, 2011
BLOOMINGTON, Ind. and SEATTLE, Wash., Nov. 14 -- The board of directors of OpenSFS, the Lustre community group in North America, has voted to fund Indiana University (IU) to develop features that will enhance the usability of the Lustre file system across wide area networks (WANs). OpenSFS is a non-profit technical organization dedicated to the promotion and development of Lustre in high performance computing (HPC) environments.
IU has been strongly committed to Lustre since 2005 when the university began planning the Data Capacitor, a high-speed, high-capacity storage system for the short- to mid-term storage of large research data sets funded by the National Science Foundation (NSF). Since then, IU has pursued and pioneered the use of Lustre as a wide area file system, deploying a production instance for scientific use in 2008. IU will be working with Lustre partner Whamcloud for testing and integration into the Lustre release tree. IU joined OpenSFS in November 2010, forming a Lustre WAN working group shortly thereafter.
"A key area of development in supercomputing systems is taking full advantage of wide area networks. IU is a clear leader in pushing Lustre WAN development, and we are excited to help support further progress in these areas," said Norman Morse, CEO of OpenSFS. "OpenSFS is deeply committed to the ongoing development of Lustre. We're pleased to fund the important work that IU is doing."
IU has proposed adding two features to the current Lustre 2.X codebase. The first is an updated and enhanced version of IU's user ID (UID) mapping code, required for any type of wide area file system that spans heterogeneous namespaces to ensure that files have proper ownership and permissions across all clients.
The second is a new flexible security model based on shared keys that will interoperate with the security code currently built into Lustre and leverage security features already available in Linux. The new features will simplify cross-domain deployment and pave the way for general adoption of Lustre as a wide area file system.
"IU's Bandwidth Challenge win was a pivotal step towards establishing Lustre as a high performance wide area file system," said Stephen Simms, manager of the High Performance File Systems group at IU. "Updating the UID mapping code will allow us to continue empowering users of geographically distributed systems and enabling new workflows. In addition, adding a new security option will help break down barriers that currently stand in the way of a shared secure Lustre across different organizations."
In September 2010, IU and Whamcloud began a partnership to further extend the capability of Lustre as a wide area file system. As part of this ongoing partnership, IU will work with Whamcloud to ensure that all new code is thoroughly tested and integrated into the Lustre release tree.
"Geographically dispersed storage is a fact of life in the supercomputing world, and tying it all together in a single namespace is a strength of Lustre that IU will help extend. This will be a real benefit to the community, and we're very excited to see OpenSFS put its full support behind IU," said Brent Gorda, CEO of Whamcloud. "This is another great example of the interesting development being done around Lustre these days. It's great news for all Lustre users."
For more information on Lustre WAN development at IU, please see http://pti.iu.edu.
Founded in 2010, OpenSFS is a non-profit California corporation focused on high-end, open-source file system technologies. The goals of the organization are to provide a forum for collaboration among entities deploying file systems on leading edge HPC systems, to communicate future requirements to open-source file system developers, and to support the release of file systems which have been designed to meet these goals. The group's initial focus is the Lustre parallel file system, which supports many of the requirements of leadership class HPC simulation environments, has a diverse development community, and is open-source software. For more information, see http://www.opensfs.org.
About Pervasive Technology Institute
Pervasive Technology Institute (PTI) at Indiana University is a world-class organization dedicated to the development and delivery of innovative information technology to advance research, education, industry, and society. Supported in part by a $15 million grant from the Lilly Endowment, Inc., PTI is built upon a spirit of collaboration and brings together researchers and technologists from a range of disciplines and organizations, including the IU School of Informatics and Computing at Bloomington, the IU Maurer School of Law, and University Information Technology Services at Indiana University. For more information, see http://pti.iu.edu.
Whamcloud, composed of HPC storage veterans and well-known Lustre experts, implements and supports Lustre solutions in HPC centers around the world. Whamcloud actively promotes the growth, stability, and vendor neutrality of Lustre. Lustre is utilized in 70% of the TOP100 supercomputing sites today and is the best technology for addressing many of the exascale issues of tomorrow. For more information, see http://www.whamcloud.com.
Source: Indiana University
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?