Introduction
Companies across all industries are facing many challenges. Emerging from the rough economy of the last few years, and facing increased global competition, virtually all organizations are in a position where they need to quickly react to market conditions and pounce on new business opportunities as they arise.
To make smarter decisions as to which markets to enter and which new opportunities to pursue, organizations now rely on much larger volumes of information. And while in the past, much of the decision-making data was historical in nature, today explosive volumes of real-time data are frequently factored in, too.
Making sense of the mountains of information and extracting relevant decision-making details now requires the use of much more sophisticated analysis, modeling, and simulations. The software that performs these tasks must work on much larger datasets than were ever used before. And the routines must run faster to keep ahead of the competition.
These factors are driving organizations to adopt high performance computing (HPC) solutions that deliver the required processing power, manage the large volumes of data, and are energy efficient.
The need for HPC
Many of the major challenges faced by organizations and government agencies today require information-based decision making. The volume of information that goes into making a decision is growing all the time, and more sophisticated algorithms are required to make sense of that information. As a result, to solve today’s increasingly complex problems, organizations require ever-more powerful computing platforms.
What’s needed is a HPC platform that has the flexibility to handle the myriad applications that are used to derive decision-making information from the raw data.
Let’s examine a few industries and see how they can benefit from HPC.
Energy: Oil and gas exploration: Many regions of the world are expected to increase their energy consumption significantly over the next decade. When conducting an exploration, analysis speed is essential when trying to determine the commercial viability of a new well or borehole. With today’s competition to find energy sources, the ultimate goal is to do better interpretation of more data in less time.
The key to achieving this analysis acceleration and higher quality decision-making is having HPC systems available to analyze and interpret exploration data. The systems must eliminate computational bottlenecks that can add days or weeks to the discovery process.
Automotive/Aerospace engineering: Computer-aided engineering (CAE) has radically changed the automotive and aerospace engineering industries. CAE has led to shorter production times and less need to develop prototypes.
There is now a growing need to support larger model sizes, more complex, multi-disciplinary models, and simulations with finer spatial and time resolution. As a result, many organizations are turning to HPC systems that deliver the required processing power to handle the more complex computations.
Banking and financial services: In today’s rapidly evolving global marketplace, financial data streams are increasing exponentially, and this massive surge of information must be analyzed and acted upon in an ever-shrinking window of opportunity.
Areas such as derivative analysis, actuarial analysis, portfolio risk measurement, and statistical analysis are driving this explosive growth and require ever-increasing compute resources to stay accurate and competitive. To support these efforts, many banking and financial services organizations are turning to HPC systems to provide the processing power, work with large datasets, and provide timely analysis that can be turned into actionable information.
Life sciences: Next-generation sequencing equipment is driving unprecedented changes in the field of life sciences. The sequencers can produce terabytes of data per experimental run and can complete each run in significantly shorter time than their predecessors.
Gaining insight from the raw output of the sequencers to enable sound decision making routinely involves a number of complex steps and the use of imaging analysis software, base calling and alignment software, and post-processing statistical analysis routines. This is an area where HPC systems are playing a major role.
How HPC helps: What’s available?
To make smarter decisions and solve today’s mix of more realistic and increasingly complex problems, organizations require ever more powerful computing platforms. The challenge is how to accommodate the broad mix of applications and explosive volumes of data, while meeting the need for speedy results using systems that are economical to manage and run.
The building blocks of a high-performance infrastructure to support these requirements include multi-core processors, application-specific accelerators, a low-latency interconnect framework, HPC-optimized software, and management tools.
A HPC solution that intelligently integrates these elements reduces complexity, reduces operational costs, drives regulatory compliance, minimizes latency, and improves risk management. Smart organizations that have turned to HPC solutions based on these technologies are realizing immediate benefits.
Examples include:
* The next-generation sequencing research at the Sanger Institute is a typical example of how HPC can accelerate life sciences R&D. Every six months, Sanger experiences a twofold increase in its compute and storage requirements as the rate of data output from the institute’s sequencers doubles. The key requirement at Sanger from HPC is analysis of the sequencing data. The rate of increase in sequencing technologies puts a lot of pressure on the HPC infrastructure.
Using an IBM System x® eX5 solution that delivers the requisite performance, the institute has been at the forefront of next generation sequence research. For more information refer to IBM eX5 improves next generation sequencing for Sanger Institute.
* A Northeast United States utility teamed with IBM to incorporate improved weather forecasting in its operations. This draws on an IBM HPC weather forecasting effort called “Deep Thunder,” which provides local, high-resolution weather predictions, customized to business applications for weather-sensitive operations up to a few days ahead of time.
The electrical utility company now can predict and prepare for weather events that can disrupt the electrical distribution network. In one recent example, the Deep Thunder forecast was available 18 hours ahead of the event and 15 hours before the National Weather Service advisory. Most importantly, Deep Thunder predicted wind gust speeds, which cause the most damage in the service area and are not included in weather prediction model.
IBM as your technology partner
To meet the HPC needs of smarter organizations, IBM offers a broad range of systems designed to meet the varied price/performance requirements of different applications.
The IBM Deep Computing portfolio includes System x and IBM Power Systems™ HPC Servers, Intelligent Cluster™, System Storage™, HPC software, services, and solutions.
HPC clusters based on IBM System x systems are x86-based servers that make use of the most current multi-core processor technology from Intel® and AMD™. The servers are available in tower, rack, or blade formats.
Servers in the System x family include:
• System x iDataPlex™: These servers are ideal for use in growing data centers that need additional performance but have space, power, and cooling constraints.
• Intelligent Cluster: Intelligent Cluster solutions include servers, storage, and interconnect technology all factory-integrated, fully tested, and ready to plug into a data center. These solutions are ideal in supporting the work of a department or line of business.
• IBM System x eX5: These servers are well-suited for applications that use and manipulate large datasets. They offer memory scaling and support for Solid State Drives (SSDs) for improved performance.
Organizations that need higher levels of computational muscle can turn to IBM Power Systems. IBM System Blue Gene® supercomputers are used to solve grand challenge problems requiring ultra scalability and energy efficiency. Blue Gene/P offers massive, near-linear scalability and leading performance. New systems based on the POWER7® processor such as the IBM Power 755 offer sustained HPC application performance needed to meet today’s smarter planet challenges. These systems support very large SMP memory, fast interconnects, and very dense packing. The higher end models are routinely used by national laboratories to conduct critical research; systems based on the POWER7 processor-based PS blades increasingly are being used in more mainstream HPC applications and are desirable for clients looking to support heterogeneous applications in a single chassis. Power blade servers and IBM Intel processor-based servers can be mixed and matched in the IBM BladeCenter® for ideal flexibility, workload optimized performance, systems management, space and power savings.
Beyond the System x and POWER server systems, IBM’s technology expertise comes into play in other ways, too. IBM System Storage®, such as the DS5000, DCS9900, and Scale Out Network Attached Storage (SONAS), are able to manage multiple petabytes of storage and up to a billion files in a single file system. IBM General Parallel File System (GPFS), available as a standalone product or as an integrated component of SONAS, helps accelerate parallel calculations on very large data sets. IBM’s software tools include parallel developer’s toolkit, engineering libraries, job schedulers, and more.
Beyond HPC systems and software, IBM has market expertise and services available to assist with smarter planet project consulting and implementations. IBM’s knowledge of applications in each domain allows the company to create industry optimized solutions.
IBM has research centers around the world that routinely team with clients, organizations and government agencies to tackle First-of-a-Kind projects. These research projects integrate IBM scientific breakthroughs and smarter planet principles with IBM expertise and solutions to help solve real-world business and societal challenges.
For more information about IBM HPC solutions designed to meet the needs of smart organizations today, visit: www.ibm.com/deepcomputing.