Posts Tagged ‘GPFS’

IBM Redefines Software Defined Storage

February 25, 2015

On Feb. 17 IBM unveiled IBM Spectrum Storage, a new storage software portfolio designed to address data storage inefficiencies by changing the economics of storage with a layer of intelligent software; in short, a software defined storage (SDS) initiative.  IBM’s new software creates an efficient data footprint that dynamically stores every bit of data at the optimal cost, helping maximize performance and ensuring security, according to the IBM announcement here.

Jared Lazarus/Feature Photo Service for IBM

Courtesy of IBM: IBM Storage GM demonstrates new Spectrum storage management dashboard

To accelerate the development of next-generation storage software, IBM included plans to invest more than $1 billion in its storage software portfolio over the next five years. The objective: extend its storage technology leadership, having recently been ranked #1 in SDS platforms for the first three quarters of 2014 by leading industry analyst firm IDC. The investment will focus on R&D of new cloud storage software, object storage, and open standard technologies including OpenStack.

“Traditional storage is inefficient in today’s world where the value of each piece of data is changing all the time,” according to Tom Rosamilia, Senior Vice President, IBM Systems, in the announcement. He went on: “IBM is revolutionizing storage with our Spectrum Storage software that helps clients to more efficiently leverage their hardware investments to extract the full business value of data.”

Two days later IBM announced another storage initiative, flash products aimed directly at, EMC. The announcement focused on two new all-flash enterprise storage solutions, FlashSystem V9000 and FlashSystem 900. Each promises industry-leading performance and efficiency, along with outstanding reliability to help lower costs and accelerate data-intensive applications. The new solutions can provide real-time analytical insights with up to 50x better performance than traditional enterprise storage, and up to 4x better capacity in less rack space than EMC XtremIO flash technology.

Driving interest in IBM Spectrum storage is research suggesting that less than 50% of storage is effectively utilized. Storage silos continue to be rampant throughout the enterprise as companies recreate islands of Hadoop-based data along with more islands of storage to support ad hoc cloud usage. Developers create yet more data silos for dev, testing, and deployment.

IBM Storage Spectrum addresses these issues and more through a SDS approach that separates storage capabilities and intelligence from the physical devices. The resulting storage is self-tuning and leverages analytics for efficiency, automation, and optimization. By capitalizing on its automatic data placement capabilities IBM reports it can meet services levels while reducing storage costs by as much as 90%.

Specifically, IBM Spectrum consists of six storage software elements:

  1. IBM Spectrum Control—analytics-driven data management to reduce costs by up to 50%
  2. IBM Spectrum Protect—optimize data protection to reduce backup costs by up to 38%
  3. IBM Spectrum Archive—fast data retention that reduces TCO for archive data by up to 90%
  4. IBM Spectrum Virtualize—virtualization of mixed environment to store up to 5x more data
  5. IBM Spectrum Accelerate—enterprise storage for cloud, which can be deployed in minutes instead of months
  6. IBM Spectrum Scale—high-performance, highly scalable storage for unstructured data

Each of these elements can be mapped back to existing IBM storage solutions.  Spectrum Accelerate, for example, uses IBM’s XIV capabilities. Spectrum virtualization is based on IBM’s San Volume Controller (SVC) technology. Spectrum Scale is based on GPFS, now called Elastic Storage, to handle file and object storage at massive scale yet within a single global name space.  Spectrum Archive, based on IBM’s LTFS, allows an organization to treat tape as a low cost, fully active tier.  In effect, with IBM Spectrum, an organization can go from flash cache to tape, all synced worldwide within a single name space.

A big part of what IBM is doing amounts to repackaging the capabilities it has built into its storage systems and proven in various products like XIV or GPFS or SVC as software components to be used as part of an SDS deployment. This raises some interesting possibilities. For instance, is it cheaper to use Spectrum Accelerate with a commodity storage array or buy the conventional XIV storage product?  The same probably could be asked of Spectrum Virtualize with SVC or Spectrum Archive with LTFS.

DancingDinosaur asked the Spectrum marketing team exactly that question.  Their response: With Accelerate you have the flexibility to size the server to the performance needs of the solution, so while the software cost remains the same regardless of the server you select. The cost of the server will vary depending on what the client needs. We will make available a sizing guide soon so each client’s situation can be modeled based on the solution requirements. In all cases it really depends on the hardware chosen vs. the (IBM) appliance. If the hardware closely matches the hardware of the appliance then costs differences will be minimal. It all depends on the price the client gets, so yes, in theory, a white box may be lower cost.

With Spectrum Accelerate (XIV), IBM continues, the client can also deploy the software on a cluster of just 3 servers (minimum) and leverage existing Ethernet networking.  This minimum configuration will be much lower cost than the minimum XIV system configuration cost. Spectrum Accelerate can also be licensed on a monthly basis, so those clients with variable needs or deploying to the cloud the client can deploy and pay for only what they need when they need it.

It is a little different for the other Spectrum offerings. DancingDinosaur will continue chasing down those details. Stay tuned. DancingDinosaur is Alan Radding, a veteran IT analyst and writer. Follow DancingDinosaur on Twitter, @mainframeblog. Follow more of his IT writing on and here.

IBM Edge2014 as Coming out Party for OpenStack

May 7, 2014

IBM didn’t invent OpenStack (Rackspace and NASA did), but IBM’s embrace of OpenStack in March 2013 as its standard for cloud computing made it a legit standard for enterprise computing. Since then IBM has made its intention to enable its product line, from the System z on down, for the OpenStack set of open source technologies.  Judging from the number of sessions at IBM Edge 2014, (Las Vegas, May 19-23 at the Venetian) that address one or another aspect of OpenStack you might think of IBM Edge2014 almost as a coming out celebration for OpenStack and enterprise cloud computing.

OpenStack is a collection of open source technologies. the goal of which is to provide a scalable computing infrastructure for both public and private clouds. As such it has become the foundation of IBM’s cloud strategy, which is another way of saying it has become what IBM sees as its future. An excellent mini-tutorial on OpenStack, IBM, and the System z can be found at mainframe-watch-Belgium here.

At IBM Edge2014 OpenStack is frequently included in sessions on storage, cloud, and storage management.  Let’s take a closer look at a few of those sessions.

IBM Storage and Cloud Technologies

Presenter Christopher Vollmar offers an overview of the IBM storage platforms that contain cloud technologies or provide a foundation for creating a private storage cloud for block and file workloads. This overview includes IBM’s SmartCloud Virtual Storage Center, SmartCloud Storage Access, Active Cloud Engine, and XIV’s Hyper-Scale as well as IBM storage products’ integration with OpenStack.

OpenStack and IBM Storage

Presenters Michael Factor and Funda Eceral explain how OpenStack is rapidly emerging as the de facto platform for Infrastructure as a Service. IBM is working fast to pin down the integration of its storage products with OpenStack. This talk presents a high level overview of OpenStack, with a focus on Cinder, the OpenStack block storage manager. They also will explain how IBM is leading the evolution of Cinder by improving the common base with features such as volume migration and ability to change the SLAs associated with the volume in the OpenStack cloud. Already IBM storage products—Storwize, XIV, DS8000, GPFS and TSM—are integrated with OpenStack, enabling self-provisioning access to features such as EasyTier or Real-time Compression via standard OpenStack interfaces. Eventually, you should expect virtually all IBM products, capabilities, and services to work with and through OpenStack.

IBM XIV and VMware: Best Practices for Your Cloud

Presenters Peter Kisich, Carlos Lizarralde argue that IBM Storage continues to lead in OpenStack integration and development. They then introduce the core services of OpenStack while focusing on how IBM storage provides open source integration with Cinder drivers for Storwize, DS8000 and XIV. They also include key examples and a demonstration of the automation and management IBM Storage offers through the OpenStack cloud platform.

IBM OpenStack Hybrid Cloud on IBM PureFlex and SoftLayer

Presenter Eric Kern explains how IBM’s latest version of OpenStack is used to showcase a hybrid cloud environment. A pair of SoftLayer servers running in IBM’s public cloud are matched with a PureFlex environment locally hosting the OpenStack controller. He covers the architecture used to set up this environment before diving into the details around deploying workloads.

Even if you never get to IBM Edge2014 it should be increasingly clear that OpenStack is quickly gaining traction and destined to emerge as central to Enterprise IT, any style of cloud computing, and IBM. OpenStack will be essential for any private, public, and hybrid cloud deployments. Come to Edge2014 and get up to speed fast on OpenStack.

Alan Radding/DancingDinosaur will be there. Look for me in the bloggers lounge between and after sessions. Also watch for upcoming posts on DancingDinosaur about OpenStack and the System z and on OpenStack on Power Systems.

Please follow DancingDinosaur on Twitter, @mainframeblog.

IBM Technical Computing Tackles Big Data

October 26, 2012

IBM Technical Computing, also referred to as high performance computing (HPC), bolstered its Platform Computing Symphony product for big data mainly by adding enterprise-ready InfoSphere BigInsights Hadoop capabilities. The Platform Symphony product now includes Apache Hadoop, map/reduce and indexing capabilities, application accelerators, and development tools. IBM’s recommended approach to simplifying and accelerating big data analytics entails the integration of Platform Symphony, General Parallel File System (GPFS), Intelligent Cluster, and DCS3700 storage.

This is not to say that IBM is leaving the traditional supercomputing and HPC market. Its Sequoia supercomputer recently topped the industry by delivering over 16 petaflops of performance.  Earlier this year it also unveiled the new LRZ SuperMUC system, built with IBM System x iDataPlex direct water cooled dx360 M4 servers encompassing more than 150,000 cores to provide a peak performance of up to three petaflops.  SuperMUC, run by Germany’s Bavarian Academy of Science’s Leibniz Supercomputing Centre, will be used to explore the frontiers of medicine, astrophysics, quantum chromodynamics, and other scientific disciplines.

But IBM is intent on broadening the scope of HPC by pushing it into mainstream business. With technical computing no longer just about supercomputers the company wants to extend technical computing to diverse industries. It already has a large presence in the petroleum, life sciences, financial services, automotive, aerospace, defense, and electronics for compute-intensive workloads. Now it is looking for new areas where a business can exploit technical computing for competitive gain.  Business analytics and big data are the first candidates that come to mind.

When it comes to big data, the Platform Symphony product already has posted some serious Hadoop benchmark results:

  • Terasort , a big data benchmark that tests the efficiency MapReduce clusters in handling very large datasets—Platform Symphony used 10x less cores
  • SWIM, a benchmark developed at UC Berkley that simulates real-world workload patterns on Hadoop clusters—Platform Symphony ran 6x faster
  • Sleep, a standard measure to compare core scheduling efficiency of MapReduce workloads—Platform Symphony came out 60x faster.

Technical computing at IBM involves System x, Power, System i, and PureFlex—just about everything except z. And it probably could run on the z too through x or p blades in the zBX.

Earlier this month IBM announced a number of technical computing enhancements including a high-performance, low-latency big data platform encompassing IBM’s Intelligent Cluster, Platform Symphony, IBM GPFS, and System Storage DCS3700. Specifically for Platform Symphony is a new low latency Hadoop multi-cluster capability that scales to 100,000 cores per application and shared memory logic for better big data application performance.

Traditionally, HPC customers coded their own software to handle the nearly mind-boggling complexity of the problems they were trying to solve. To expand technical computing to mainstream business, IBM has lined up a set of ISVs to provide packaged applications covering CAE, Life Science, EDA, and more. These include Rogue Wave, ScaleMP, Ansys, Altair, Accelrys, Cadence, Synopsys, and others.

IBM also introduced the new Flex System HPC Starter Configuration, a hybrid system that can handle both POWER7 and System x.  The starter config includes the Flex Enterprise Chassis, an Infiniband (IB) chassis switch, Power7 compute node, and an IB expansion card for Power or x86 nodes. Platform Computing software handles workload management and optimizes resources. IBM describes it as a high density, price/performance offering but hasn’t publicly provided any pricing. Still, it should speed time to HPC.

As technical computing goes mainstream it will increasingly focus on big data and Hadoop.  Compute-intensive, scientific-oriented companies already do HPC. The newcomers want to use big data techniques to identify fraud, reduce customer churn, make sense of customer sentiment, and similar activities associated with big data. Today that calls for Hadoop which has become the de facto standard for big data, although that may change going forward as a growing set of alternatives to Hadoop gain traction.


Get every new post delivered to your Inbox.

Join 719 other followers

%d bloggers like this: