High Performance Computing Breaks New Ground
- SUSE Partners for High Performance Computing
- SUSE Customer Success Stories with SUSE Linux Enterprise Server
High Performance Computing (HPC) is used to solve the most demanding computational and data-intensive problems. Due to the evolution of both lower cost hardware and enterprise class Linux distributions, the HPC market is rapidly changing. It is moving away from RISC/UNIX symmetric multiprocessor (SMP) servers and proprietary cluster platforms, and moving toward Linux industry standard servers and clusters. Linux has become the cluster operating system of choice, thanks to its scalability and performance capabilities, its similarity to UNIX, and the wide variety of open-source software and development tools available.
This evolution has allowed the cost of HPC systems to be dramatically reduced. And with compute power increasing a thousand times in just a few years, many more companies can now utilize the power of "supercomputers" in the form of an HPC Linux cluster.
If you are under pressure to reduce server floor space usage and more effectively manage your organization's insatiable appetite for increased performance, new and improved services, and enhanced productivity, you're not alone. HPC Linux clusters are an ideal solution to these challenges.
HPC is no longer confined to the traditional applications and workloads run by technical computing groups. Commercial customers are now adopting HPC for applications such as financial analysis and portfolio management, digital security and surveillance, as well as decision-support computing.
HPC systems are also being utilized for more and more business uses, such as data warehouses, line-of-business (LOB) applications and transaction processing. And while HPC has been primarily limited to large enterprises, R&D firms, and academic institutions in the past, that is changing. There is now also a broad swath of mid-market companies adopting HPC. This is due to the availability of affordable and open solutions, which supplant the costly proprietary solutions of the past. All of these factors are contributing to a logical progression: HPC is maturing from "high performance" to "high productivity."
SUSE Linux Enterprise Server—Best Choice for High Performance Computing
SUSE Linux Enterprise Server is synonymous with high-performance Linux running on 64-bit and mainframe systems. Since 1993, SUSE engineers have made significant contributions to the advancement and tuning of the Linux kernel, and key kernel-related performance technologies.
Today, multi-core processors are setting the stage for new market dynamics in high performance computing. That's why enterprises worldwide are turning to SUSE Linux Enterprise Server as their preferred operating system in this area.
With advanced memory management, new processor support, unmatched performance on systems with multicore processors, Native POSIX Thread Library (NPTL) and advanced multi-pathing and I/O capabilities, SUSE Linux Enterprise Server is the clear choice. It outperforms RISC/UNIX systems in performance and scalability for large-scale server deployments, both on commodity blades and servers, and on high-end mainframes.
SUSE Linux Enterprise Server has been designed to handle mission-critical workloads in the data center. It ships with file systems perfectly suited for large scale environments. SUSE Linux Enterprise Server also includes OpenFabrics Enterprise Distribution (OFED) 1.4, the most current open source InfiniBand software stack offered by the OpenFabrics Alliance. OFED 1.4 includes 10 Gigabit Ethernet drivers and MPICH2 MPI libraries. With SUSE Linux Enterprise Server, you can maximize your investment in blade server infrastructure with higher density configurations. This allows you to reduce the time, cost and complexity of delivering high-bandwidth connections for high-performance computing.
By looking at the Top500.org web site you can see that SUSE Linux Enterprise Server is the operating system of choice on the world's largest HPC "super computers" in use today. Of the Top 500 supercomputers, more than 85% are running on Linux–and you will always find several top 10 supercomputers running SUSE Linux Enterprise Server. For example, Europe's most powerful supercomputer, SuperMUC, ranked number four in the June 2012 TOP 500 list, is powered by SUSE Linux Enterprise Server. SuperMUC is a System X iDataPlex from IBM and is equipped with more than 155,000 processor cores, which deliver an aggregate peak performance of more than 3 Petaflops of computing power.
To help customers reduce the complexity and risk associated with buying an HPC cluster solution, we participate in the Intel Cluster Ready program. Developed in conjunction with hardware and software vendors, the Intel Cluster Ready program is designed to simplify purchasing, deployment, and management of HPC clusters. Now you can rely on SUSE Linux Enterprise Server powering many of the certified Intel Cluster Ready systems.
Real Time Capabilities
As HPC systems began to push for faster and faster response times, especially in some critical need environments, end users began demanding real time response for their compute needs. Real time Linux was driven first by military and scientific needs. Today, however, enterprises such as banking and Wall Street trading demand real time support as well. Real time capabilities are also a big interest for organizations that seek to eliminate jitter to gain better performance. We have developed real time add-ons for SUSE Linux Enterprise that allow for real time response for the end users that need this level of support. SUSE Linux Enterprise Real Time is an industry-standard real-time Linux operating system that has been specifically engineered to reduce the latency and increase the predictability and reliability of time-sensitive mission-critical applications.
The Importance of Interoperability
In many organizations, the boundaries between HPC and enterprise IT organizations have all but disappeared. HPC systems, ranging from a few nodes to clusters boasting thousands of processors, must now operate within the same environments that measure not only performance, but also reliability, support, total cost of ownership and productivity.
Leading-edge HPC groups and enterprise IT organizations have converged on the same path—one that is leading them to the data center of the future. This data center includes a diverse yet interoperable platform environment—with a true coexistence of proprietary and open systems. Key applications may dictate platform and architecture requirements and force organizations to maintain both proprietary and open source HPC environments. Interoperability in an HPC environment is critical to reducing costs, future-proofing HPC infrastructure investments and improving productivity. SUSE and Microsoft are bringing unparalleled productivity to mixed HPC cluster environments through Linux and Windows interoperability.
Because HPC clusters represent a significant investment, maximizing the benefits of your current assets can result in huge savings. With a dual-boot capability—using the same cluster for two operating systems—you gain the flexibility of an additional cluster without having to invest in further hardware.