The new intelligent resource and workload management software, called IBM Spectrum Computing, is designed to make it easier for organizations to extract fullvalue from data to accelerate performance-intensive analytics or machine learning. This technology can be used across industries to help sequence genomes for improved cancer treatment, or for engineers to design a championship-winning Formula One race car or bankers to personalize financial services to attract new customers.

The IBM Spectrum Computing platform offers new cognitive, resource-aware scheduling policies that help increase the utilization of existing compute resources, controlling costs while speeding results for high performance computing, big data analytics, new generation applications and open source frameworks, such as Hadoop and Apache Spark. Here are five things to know about IBM High Performance Computing solutions:

1. Open systems designed for the intersection of HPC and Big Data.
   Although HPC and big data analytics are converging, traditional HPC clusters are built for another era—meaning they are designed for data or compute, but not both. This leads to the inability to deliver adequate performance and scalability, along with I/O bottlenecks and network latency when moving large datasets—all of which slows down real-time insights. IBM HPC clusters deliver more when they are built with Power Systems—so you can easily handle demanding workloads and high performance data analytics.
    
2. Utilizes new techniques for HPC with IBM PowerAI.
   Traditional HPC algorithms are no longer able to feasibly solve every problem without massive deployments. New approaches are required. Machine learning and artificial intelligence can now be rapidly deployed for all kinds of workloads with PowerAI. IBM is a leader in machine learning, deep-learning, and AI techniques applied to computing. Just ask Watson and you’ll see what we’re talking about.
   
   

A simplified view of an HPC system

3. Solution built on IBM Power Systems HPC Cluster.
   An integrated offering combining high performance compute nodes, low latency interconnect fabric, optional high performance parallel storage, and optional system software.  IBM Power Systems high performance computing servers are deployed in many of the largest clusters in the world. Configurable into highly scalable Linux clusters, Power Systems offer extreme performance for demanding workloads such as genomics, finance, computational chemistry, oil and gas exploration, and high performance data analytics.
    
4. IBM Power Systems servers for high performance computing.
   The IBM Power System S822LC is built on industry standards and incorporates innovation from the OpenPOWER Foundation ecosystem: including up to 2 NVIDIA Tesla GPU Accelerators and Mellanox InfiniBand. The Power S822LC delivers faster time to insight by pairing the built-for-big-data architecture of POWER8 and accelerator performance. Also, available without GPU accelerators as IBM Power System S822LC for commercial computing.
    
5. Delivers parallel storage with IBM Spectrum Scale.
   Implementing the unique advantages of IBM Spectrum Scale (formerly GPFS), IBM Elastic Storage Server is a storage solution that provides persistent performance at any scale. It ensures fast access and availability of the right data, at the right time, across clusters.

The S822LC for HPC server with water-cooled GPU           

IBM HPC solutions are built for data-centric computing, and delivered with integration expertise targeting performance optimization at the workflow level. Data-centric design minimizes data motion, enables compute capabilities across the system stack, and provides a modular, scalable architecture that is optimized for HPC.

For more information on the IBM High-Performance Computing solutions, refer to the following:

IBM POWER8 High-Performance Computing Guide: IBM Power System S822LC (8335-GTB) IBM Redbooks publication
IBM High Performance Computing Solutions 
IBM Power Systems Home Page
 

Joseph T Apuzzo
Markus Hilger
Desnes Augusto Nunes Rosario
Alexander Pozdneev
Dino Quintero