Big Data Servers offer alternative to commodity servers.

Press Release Summary:



Built on OpenPOWER stack, IBM Power S824L Servers give clients ability to run data-intensive tasks on POWER8 processor while offloading other compute-intensive Big Data workloads to GPU accelerators, which are capable of running millions of data computations in parallel. With optimized computing performance, banks can better analyze risk, energy companies can precisely locate oil reserves, and scientists can more quickly identify cures for diseases.



Original Press Release:



IBM Provides Clients Superior Alternative to x86-Based Commodity Servers



OpenPOWER-Based Systems Deliver Nearly 20 Percent Better Price/Performance than x86-Based Servers



ARMONK, N.Y. -- IBM (NYSE: IBM) today introduced a new range of systems capable of handling massive amounts of computational data faster at nearly 20 percent better price/performance(1,2) than comparable Intel Xeon v3 Processor-based systems - providing clients a superior alternative to closed, commodity-based data center servers.



The vastness of Big Data - of the 2.5 quintillion bytes of data generated on the planet every day - provides a challenge for organizations to build powerful technology infrastructures to gain actionable insights from this data. The commodity servers used today by most organizations are built on proprietary processor technology and are increasingly stretched to their limits by workloads related to Big Data, cloud and mobile demands. In contrast, IBM is designing a new data centric approach to systems that leverages the building blocks of the OpenPOWER Foundation. Formed in late 2013, the OpenPOWER Foundation has grown quickly from 5 founders to 59 members - all working together to leverage the IBM POWER processor's open architecture for broad industry innovation.



"Our open innovation business model and approach to OpenPOWER will disrupt technology providers that offer closed, proprietary solutions produced within the walls of one company," said Doug Balog, General Manager, Power Systems, IBM Systems & Technology Group. "Today's announcement provides clients with a greater choice to help them gain the fastest, deepest insights to solve their business problems."



The new IBM Power S824L servers are built on IBM's POWER8 processor, the world's first processor optimized for the most demanding Big Data workloads(3). The new systems tightly integrate IBM and other OpenPOWER member technologies, including NVIDIA's GPU accelerator technology for the first time, to unleash incredible computing performance to help enable banks to better analyze risk, energy companies to more precisely locate oil reserves, and scientists to more quickly identify cures for diseases.



Built on the OpenPOWER stack(4), the Power S824L systems provide clients the ability to run data-intensive tasks on the POWER8 processor while offloading other compute-intensive Big Data workloads to GPU accelerators which are capable of running millions of data computations in parallel and are designed to significantly speed up compute-intensive applications.



To take advantage of GPU acceleration on Power Systems, IBM will be optimizing IBM Big Data enterprise applications, including the IBM DB2 database software with BLU Acceleration. Additionally, IBM is working to optimize Power versions of widely used GPU-accelerated applications for bioinformatics, defense, finance, molecular dynamics, weather modeling - including SOAP3, NAMD, GROMACS, FFTW library, and Quantum Espresso.



Future versions of IBM Power Systems will feature NVIDIA NVLink technology, eliminating the need to transfer data between the CPU and GPUs over the PCI Express interface. This will enable NVIDIA GPUs to access IBM POWER CPU memory at its full bandwidth, improving performance for numerous enterprise applications. Such systems are expected to be available beginning in 2016.



POWER8 Portfolio Expands



In addition to the GPU-accelerated Power S824L offering, other additions to IBM's POWER8 processor-based Power Systems portfolio and solutions that complement the lineup include:



--  IBM Data Engine for NoSQL: Bringing together innovation from OpenPOWER members Redis Labs, Canonical, Altera and IBM, this solution enables a significantly lower cost basis for deploying NoSQL data stores. The solution combines Redis Labs software, Canonical's Ubuntu Linux operating system, Altera's FPGAs and IBM's unique CAPI-enabled Power System S822L with IBM's FlashSystem 840. Simplicity is achieved through server consolidation, enabling one POWER8 server to be used instead of 24 Intel-based servers for a well-sized NoSQL store.(5) A many-fold reduction in infrastructure cost can be realized through this smaller footprint and associated energy cost reductions. Clients can realize new opportunities and grow far larger NoSQL stores by leveraging the full 40TB of Flash within a single rack.



--  IBM Data Engine for Analytics - Power Systems Edition: Combining POWER8 Scale-out systems with IBM's flash-based elastic storage technology and Platform Computing software, this new analytics appliance offering can use a third of the storage infrastructure as required by an x86-based solution.(6)



--  Power Enterprise Systems: Built with the fastest performing core and fastest performing high-end server chip in the industry(7), the new Power Enterprise Systems expand the lineup from the POWER8 based scale-out servers introduced in April and are designed and optimized for the demands of enterprise data and computing environments. The new Power E870 and Power E880 Systems are the highest performance 8-socket systems in the industry with support for up to 1,000 VMs per system(8). With choices of 32, 40 or 48 processor cores and up to 4 TB per compute node, these systems are designed for modular efficiency and seamless growth and will grow to a massive 192 cores with more than 1500 threads of compute power and 16 TB of memory, enabling them to handle the most data intensive, mission critical applications in the industry. Capable of running AIX, IBM i, and Linux operating systems, the new Power Enterprise Systems are designed for clients to confidently support the most demanding, mission-critical environments.



--  Power Enterprise Pools: Delivering a powerful foundation for private, public, and hybrid cloud infrastructure, Enterprise Pools offer extreme flexibility and instant response to changes in workload or infrastructure. They enable clients to move resources across a pool of systems to match workload demand, help ensure system availability, improve efficiency, and protect investments while transitioning to POWER8.



With general availability planned for October 31, the new IBM Power systems and solutions come in variety of configurations and service arrangements.



Financing Available

IBM Global Financing provides clients with affordable options to acquire the latest in IBM technology and solutions. Offerings available include 0% financing for the Power S824L servers, customized financing structures, and programs to migrate to the new Power enterprise servers.



IBM Global Financing is the world's largest IT captive financier and can help enable credit-qualified clients to transform their business. IBM's flexible payment options can accelerate the acquisition of transformational projects with total solution financing, including IBM and non-IBM products and services. IBM's IT financing specialists have serviced clients in over 60 countries in 33 years.



To learn more about the IBM Power Systems launched today visit: http://www.ibm.com/systems/power/announcement.html



To learn more about the OpenPOWER Foundation visit www.openpowerfoundation.org



Contact:

Kristin Bryson

IBM

+1 (203) 241-9190

kabryson@us.ibm.com



Contact:

Ciri Haugh

IBM

+1 (512) 286-7009

cehaugh@us.ibm.com



(1) Performance is based on an estimate of published SPECcpu2006 (SPECfp_rate2006) results as of September 26, 2014 http://www.specbench.org/. 



(2) IBM pricing based on: IBM Power Systems S824L (2x12 cores/96 threads/3.02GHz) configured with 192 GB Memory (8GB/core) with Ubuntu Advantage Linux. x86 pricing based on: HP DL380p with Intel E5-2699 v3 (2x18 cores/36 threads/2.3GHz) configured with 288 GB Memory (8GB/core) with Red Hat Enterprise Linux.   



(3) By Big Data, IBM means handling both extraordinary large volumes of structured (relational databases) and unstructured (noSQL, Map Reduce) data from which customers derive analytics and insight. The POWER8 processor is the first processor with differentiated capabilities designed to handle both structured and unstructured data. Such capabilities include: the CAPI (Coherent Accelerator Processor Interface) Architecture with key data capabilities e.g.: a). CAPI Flash Access Efficiency, b). Storage reduction via CAPI Attached Compression Accelerator, and c). Throughput and latency advantage of CAPI Attached Mellanox RDMA Fabric. DBMA (dynamic balanced memory architecture) in the form of key capabilities, e.g. a). Internal Processor Data Flow b). Memory Bandwidth advantage c). Cache capacity advantage



(4) OpenPOWER stack consists of an open set of hardware and software specifications maintained by the OpenPOWER Foundation, based on the IBM POWER architecture.



(5) 24:1 system consolidation ratio (12:1 rack density improvement) based on a single IBM S824, (24 cores, POWER8 3.5 GHz), 256GB RAM, AIX 7.1 with 40 TB memory based Flash replacing 24 HP DL380p, 24 cores, E5-2697 v2 2.7 GHz), 256GB RAM, SuSE Linux 11SP3 . Inbound network limits performance to 1M IOPs in both scenarios, equal capacity (#user, data) in both cases.



(6) Over 3X less storage infrastructure for Power as compared to a typical Intel Hadoop based solution.  Based on IBM internal testing, clients actual results may vary. The IBM POWER8 servers with GPFS file system used 116 (3.5" HDS) disk drives.  The Intel servers with HDFS file system required 280 (3.5" HDS) disk drives.  Additional storage savings comes from eliminating data duplication during Extraction, Transformation, and Loads (ETL) processing.



(7) Based on benchmarks and internal testing.



(8) Based on benchmarks and internal testing.

All Topics