- Intersect360 Research User Site Census: Applications, July 2010
- Intersect360 Research Traditional HPC Total Market Forecast: 2010 to 2014, June 2010
- Intersect360 Research Site Census Report – Storage, February 2010
- HPC User Site Census: Interconnects/Networks, January 2010
- HPC User Site Census: Processors, December 2009
- Cloud Computing Opportunities in HPC, December 2009
- HPC User Site Census: Systems, September 2009
- Traditional HPC Total Market Forecast: 2009 to 2013 , September 2009
- HPC Site Budget Allocation Map: Economic Sectors, July 2009
- Traditional HPC 2008 Market Model: May 2009, May 2009
- 2008 Technical HPC Server Market Review, April 2009
- Risks and Rewards: SGI Sells Out to Rackable, April 2009
- InfiniBand: Increases in Speed, Usage, Competition, March 2009
- HPC User Site Census: Application Software, March 2009
- Platform Computing in the Catbird Seat, February 2009
- 2008 Top Ten Study Results From Tabor Research, February 2009
- Preliminary 2008 Market Review, January 2009, January 2009
Intersect360 Research Site Census Report – Storage
by Addison Snell, Christopher G. Willard, Ph.D., Laura Segervallfor Intersect360 Research
February 2010
This fourth report in our Site Census series provides an examination of the storage characteristics and capacities found at a sample of HPC user sites. We surveyed a broad range of users about their current computer system installations, storage systems, networks, middleware, and software supporting these computer installations. The initial three reports, HPC User Site Census: Systems, HPC User Site Census: Processors, and HPC User Site Census: Interconnects/Networks focused on server suppliers, server node characteristics, processor-related trends, and system interconnect and network utilization.1
Our goal in this analysis of storage systems is to examine storage usage within the HPC user communities and explore how this usage varies based on categories such as storage capacity and its location, storage supplier, storage network, etc.
Key findings of the survey include the following:
- Approximately 47% of the total maximum available storage at respondents' sites resides on compute servers. Storage available to each node (referred to as node-level storage) represents 7% of the capacity, and storage available to the server (referred to as system-level storage) accounts for 40% of the total available storage. The remaining 53% of storage is found at the site level, generally on NAS or SAN systems. On average, 450TB of storage resides at the site level on a storage system.
- About 75% of the 147 sites have at least one site-level storage system installed. No vendor dominates the storage system market for HPC sites. IBM has the largest share with 10.7% but is closely followed by EMC and Sun with 10.1% each. However, “in-house” and “generic” solutions combine to account for 10.7% of the storage systems installed. We see this last value as reflecting the commodity nature of storage components and the availability of open software for storage systems.
- NAS (Networked-Attached Storage) and SAN (Storage Area Network) had almost equal representation in the surveyed HPC sites, with 39% SAN and 34% NAS. Commercial sites are more likely to have a NAS storage system while academic sites are more likely to have a SAN storage system.
- The majority of NAS storage systems are connected using 1 Gigabit Ethernet. Only 8% of the systems were connected to a 10 Gigabit Ethernet network.
- Most storage management software (38%) in use by these sites was provided by the storage system vendor. Very little penetration by add-on storage management suppliers was reported.
1Future reports will deal with operating systems and software individually.
©2010 Intersect 360 Inc. All rights reserved.
| HPC User Site Census: Systems |
» info@intersect360.com
» 408-306-2926


