The University of Texas at San Antonio (UTSA) is expected to become one of the nation’s premier research universities, achieving that status over a relatively short interval by focusing on academic excellence, faculty and student led research, and by extending their reach internationally. Today, UTSA is the second-largest university in Texas and continues to develop capabilities allowing it to become recognized as a first-rank research institution.
UTSA selected AMD’s SeaMicro SM15000-OP technology to allow creation of a computing cloud that will help the institution achieve its objectives of enabling breakthrough scientific discoveries, attracting superior faculty and students, Increasing competitiveness for obtaining research funding, and expanding access to cloud computing resources.
AMD’s SeaMicro SM15000 Fabric Compute Systems family delivers a revolutionary data center platform that brings together computing, networking, and storage in a single 10 RU energy efficient system. The SeaMicro SM15000-OP servers are powered by a combined 1,024 AMD Opteron processor cores in 20 rack units (occupying 35 inches of space), and serve as the foundation for a computing cloud, powered by OpenStack. AMD notes that state-of-the-art cloud computing is becoming keystone infrastructure for advancing superior current and future research, noting that research projects today increasingly require inter- disciplinary collaboration, huge data storage capacity, and advanced computational capabilities. Procuring and managing computing and storage infrastructure creates overhead that takes up valuable time and energy from a research team’s staff. With cloud computing, researchers are freed from the burden of managing IT equipment and can focus on their research. This new deployment allows the broader UTSA community to realize the benefits of cloud computing by making it more widely available and easier to use.
Rackspace Private Cloud Certification Provides Peace of Mind
AMD reports that the process of evaluating and selecting the SeaMicro SM15000 for the new computing cloud was completed in just one week, thanks to the server’s certifications for Rackspace Private Cloud for OpenStack Mass Compute (Nova) and OpenStack Object Storage (Swift). The certification and pre-tested architectures provided UTSA with out-of-the-box solutions tested and proven to work for real life deployments. The ability to provision and configure computing and storage resources on demand for any research project was expected to eliminate weeks, sometimes months of setup and optimization that will free research teams from the time-consuming task of identifying, procuring, installing, and managing their own IT equipment.
AMD claims that its SeaMicro SM15000-OP system is the highest-density server on the market, offering industry-leading energy efficiency. In 10 rack units, it links 512 compute cores from the next generation AMD Opteron processor, 16 x 10 Gigabits per second of networking I/O, more than five petabytes of Freedom Fabric Storage with a 1.28 terabyte high-performance.
“World-class computing advances UTSA’s cutting edge research and discovery of new knowledge,” says C. Mauli Agrawal, Dean of the University of Texas at San Antonio College of Engineering in a release. “As an emerging research university, this project supports our mission of providing world- class education, outstanding research, and economic contributions to the region.”
Computing and storage requirements for research are wide-ranging. Some applications are compute intensive and require little storage, while others have the exact opposite requirements. Traditionally, says AMD, the challenge has been that expanding a data center required deploying more servers, which meant expanding compute, storage, and networking. However, the SM15000 server overcomes this limitation by allowing storage, compute, and networking to all be expanded or upgraded independently. This provides a tremendous advantage to UTSA since their computing and storage needs vary from project to project.
Fabric Servers can be provisioned on-demand through a unified management console and right-sized with needed storage and network resources to best meet application demands. Touted as “the industry’s first data center in a box,” AMD’s SeaMicro SM15000 server is built around the industry’s first and only second-generation fabric, called SeaMicro Freedom Fabric, claimed to be the only fabric technology designed and optimized to work with Central Processor Units (CPUs) that have both large and small cores, as well as x86 and ARM CPUs. AMD’s SeaMicro server product family currently supports the next generation AMD Opteron (“Piledriver”) processor, Intel Xeon E3-1260L (“Sandy Bridge”) and E3-1265Lv2 (“Ivy Bridge”) and Intel Atom N570 processors.
High capacity, external storage is provided by Freedom Fabric Storage technology, which enables a single SM15000-OP system to provide more than five petabytes of storage with 64 servers in a single ten-rack unit (17.5 inches tall) system. Once these disks are interconnected with the fabric, they are seen and shared by all servers in the system. This approach provides benefits typically provided by expensive and complex solutions such as network attached storage (NAS) and storage area networking (SAN), with the simplicity and low cost of direct attached storage. Whether the project is to undertake a large scale study of proteins, simulate high throughput biochemical systems, or analyze computational fluid dynamics, the SM15000 server can supply a powerful and flexible cloud computing platform.
“We envision the computing cloud built on AMD’s SeaMicro servers and powered by OpenStack to change the paradigm of research at UTSA. It will become a critical service by eliminating the need for each team to buy and manage its own infrastructure, and it will foster collaboration and sharing by making the computing resources and data accessible from anywhere,” Jeff Prevost, Cloud Technology Institute comments in the AMD release.
More information about AMD and the SeaMicro family of high density, low power servers can be found at: