Posts Tagged ‘POWER8’

IBM Shows Off POWER and NVIDIA GPU Setting High Performance Record 

May 4, 2017

The record achievement used 60 Power processors and 120 GPU accelerators to shatter the previous supercomputer record, which used over a 700,000 processors. The results point to how dramatically the capabilities of high performance computing (HPC) has increase while the cost of HPC systems has declined. Or put another way: the effort demonstrates the ability of NVIDIA GPUs to simulate one billion cell models in a fraction of the time, while delivering 10x the performance and efficiency.

Courtesy of IBM: Takes a lot of processing to take you into a tornado

In short, the combined success of IBM and NVIDIA puts the power of cognitive computing within the reach of mainstream enterprise data centers. Specifically the project performed reservoir modeling to predict the flow of oil, water, and natural gas in the subsurface of the earth before they attempt to extract the maximum oil in the most efficient way. The effort, in this case, involved a billion-cell simulation, which took just 92 minutes using 30 for HPC servers equipped with 60 POWER processors and 120 NVIDIA Tesla P100 GPU accelerators.

“This calculation is a very salient demonstration of the computational capability and density of solution that GPUs offer. That speed lets reservoir engineers run more models and ‘what-if’ scenarios than previously,” according to Vincent Natoli, President of Stone Ridge Technology, as quoted in the IBM announcement. “By increasing compute performance and efficiency by more than an order of magnitude, we’re democratizing HPC for the reservoir simulation community,” he added.

“The milestone calculation illuminates the advantages of the IBM POWER architecture for data-intensive and cognitive workloads.” said Sumit Gupta, IBM Vice President, High Performance Computing, AI & Analytics in the IBM announcement. “By running Stone Ridge’s ECHELON on IBM Power Systems, users can achieve faster run-times using a fraction of the hardware.” Gupta continued. The previous record used more than 700,000 processors in a supercomputer installation that occupies nearly half a football field while Stone Ridge did this calculation on two racks of IBM Power Systems that could fit in the space of half a ping-pong table.”

This latest advance challenges perceived misconceptions that GPUs could not be efficient on complex application codes like reservoir simulation and are better suited to simple, more naturally parallel applications such as seismic imaging. The scale, speed, and efficiency of the reported result disprove this misconception. The milestone calculation with a relatively small server infrastructure enables small and medium-size oil and energy companies to take advantage of computer-based reservoir modeling and optimize production from their asset portfolio.

Billion cell simulations in the industry are rare in practice, but the calculation was accomplished to highlight the performance differences between new fully GPU-based codes like the ECHELON reservoir simulator and equivalent legacy CPU codes. ECHELON scales from the cluster to the workstation and while it can simulate a billion cells on 30 servers, it can also run smaller models on a single server or even on a single NVIDIA P100 board in a desktop workstation, the latter two use cases being more in the sweet spot for the energy industry, according to IBM.

As importantly, the company notes, this latest breakthrough showcases the ability of IBM Power Systems with NVIDIA GPUs to achieve similar performance leaps in other fields such as computational fluid dynamics, structural mechanics, climate modeling, and others that are widely used throughout the manufacturing and scientific community. By taking advantage of POWER and GPUs organizations can literally do more with less, which often is an executive’s impossible demand.

DancingDinosaur is Alan Radding, a veteran information technology analyst, writer, and ghost-writer. Please follow DancingDinosaur on Twitter, @mainframeblog. See more of his IT writing at technologywriter.com and here.

 

IBM Gets Serious About Open Data Science (ODS) with Anaconda

April 21, 2017

As IBM rapidly ramps up cognitive systems in various forms, its two remaining platforms, z System and POWER, get more and more interesting. This week IBM announced it was bringing the Anaconda Open Data Science (ODS) platform to its Cognitive Systems and PowerAI.

Anaconda, Courtesy Pinterest

Specifically, Anaconda will integrate with the PowerAI software distribution for machine learning (ML) and deep learning (DL). The goal: make it simple and fast to take advantage of Power performance and GPU optimization for data-intensive cognitive workloads.

“Anaconda on IBM Cognitive Systems empowers developers and data scientists to build and deploy deep learning applications that are ready to scale,” said Bob Picciano, senior vice president of IBM Cognitive Systems. Added Travis Oliphant, co-founder and chief data scientist, Continuum Analytics, which introduced the Anaconda platform: “By optimizing Anaconda on Power, developers will also gain access to the libraries in the PowerAI Platform for exploration and deployment in Anaconda Enterprise.”

With more than 16 million downloads to date, Anaconda has emerged as the Open Data Science platform leader. It is empowering leading businesses across industries worldwide with tools to identify patterns in data, uncover key insights, and transform basic data into the intelligence required to solve the world’s most challenging problems.

As one of the fastest growing fields of AI, DL makes it possible to process enormous datasets with millions or even billions of elements and extract useful predictive models. DL is transforming the businesses of leading consumer Web and mobile application companies, and it is catching on with more traditional business.

IBM developed PowerAI to accelerate enterprise adoption of open-source ML and DL frameworks used to build cognitive applications. PowerAI promises to reduce the complexity and risk of deploying these open source frameworks for enterprises on the Power architecture and is tuned for high performance, according to IBM. With PowerAI, organizations also can realize the benefit of enterprise support on IBM Cognitive Systems HPC platforms used in the most demanding commercial, academic, and hyperscale environments

For POWER shops getting into Anaconda, which is based on Python, is straightforward. You need a Power8 with IBM GPU hardware or a Power8 combined with a Nvidia GPU, in effect a Minsky machine. It’s essentially a developer’s tool although ODS proponents see it more broadly, bridging the gap between traditional IT and lines of business, shifting traditional roles, and creating new roles. In short, they envision scientists, mathematicians, engineers, business people, and more getting involved in ODS.

The technology is designed to run on the user’s desktop but is packaged and priced as a cloud subscription with a base package of 20 users. User licenses range from $500 per year to $30,000 per year depending on which bells and whistles you include. The number of options is pretty extensive.

According to IBM, this started with PowerAI to accelerate enterprise adoption of open-source ML/DL learning frameworks used to build cognitive applications. Overall, the open Anaconda platform brings capabilities for large-scale data processing, predictive analytics, and scientific computing to simplify package management and deployment. Developers using open source ML/DL components can use Power as the deployment platform and take advantage of Power optimization and GPU differentiation for NVIDIA.

Not to be left out, IBM noted growing support for the OpenPOWER Foundation, which recently announced the OpenPOWER Machine Learning Work Group (OPMLWG). The new OPMLWG includes members like Google, NVIDIA and Mellanox to provide a forum for collaboration that will help define frameworks for the productive development and deployment of ML solutions using OpenPOWER ecosystem technology. The foundation has also surpassed 300-members, with new participants such as Kinetica, Red Hat, and Toshiba. For traditional enterprise data centers, the future increasingly is pointing toward cognitive in one form or another.

DancingDinosaur is Alan Radding, a veteran information technology analyst, writer, and ghost-writer. Please follow DancingDinosaur on Twitter, @mainframeblog. See more of his IT writing at technologywriter.com and here.

 

Open POWER-Open Compute-POWER9 at Open Compute Summit

March 16, 2017

Bryan Talik, President, OpenPOWER Foundation provides a detailed rundown on the action at the Open Compute  Summit held last week in Santa Clara. After weeks of writing about Cognitive, Machine Learning, Blockchain, and even quantum computing, it is a nice shift to conventional computing platforms that should still be viewed as strategic initiatives.

The OpenPOWER, Open Compute gospel was filling the air in Santa Clara.  As reported, Andy Walsh, Xilinx Director of Strategic Market Development and OpenPOWER Foundation Board member explained, “We very much support open standards and the broad innovation they foster. Open Compute and OpenPOWER are catalysts in enabling new data center capabilities in computing, storage, and networking.”

Added Adam Smith, CEO of Alpha Data:  “Open standards and communities lead to rapid innovation…We are proud to support the latest advances of OpenPOWER accelerator technology featuring Xilinx FPGAs.”

John Zannos, Canonical OpenPOWER Board Chair chimed in: For 2017, the OpenPOWER Board approved four areas of focus that include machine learning/AI, database and analytics, cloud applications and containers. The strategy for 2017 also includes plans to extend OpenPOWER’s reach worldwide and promote technical innovations at various academic labs and in industry. Finally, the group plans to open additional application-oriented workgroups to further technical solutions that benefits specific application areas.

Not surprisingly, some members even see collaboration as the key to satisfying the performance demands that the computing market craves. “The computing industry is at an inflection point between conventional processing and specialized processing,” according to Aaron Sullivan, distinguished engineer at Rackspace. “

To satisfy this shift, Rackspace and Google announced an OCP-OpenPOWER server platform last year, codenamed Zaius and Barreleye G2.  It is based on POWER9. At the OCP Summit, both companies put on a public display of the two products.

This server platform promises to improve the performance, bandwidth, and power consumption demands for emerging applications that leverage machine learning, cognitive systems, real-time analytics and big data platforms. The OCP players plan to continue their work alongside Google, OpenPOWER, OpenCAPI, and other Zaius project members.

Andy Walsh, Xilinx Director of Strategic Market Development and OpenPOWER Foundation Board member explains: “We very much support open standards and the broad innovation they foster. Open Compute and OpenPOWER are catalysts in enabling new data center capabilities in computing, storage, and networking.”

This Zaius and Barreleye G@ server platforms promise to advance the performance, bandwidth and power consumption demands for emerging applications that leverage the latest advanced technologies. These latest technologies are none other than the strategic imperatives–cognitive, machine learning, real-time analytics–IBM has been repeating like a mantra for months.

Open Compute Projects also were displayed at the Summit. Specifically, as reported: Google and Rackspace, published the Zaius specification to Open Compute in October 2016, and had engineers to explain the specification process and to give attendees a starting point for their own server design.

Other Open Compute members, reportedly, also were there. Inventec showed a POWER9 OpenPOWER server based on the Zaius server specification. Mellanox showcased ConnectX-5, its next generation networking adaptor that features 100Gb/s Infiniband and Ethernet. This adaptor supports PCIe Gen4 and CAPI2.0, providing a higher performance and a coherent connection to the POWER9 processor vs. PCIe Gen3.

Others, reported by Talik, included Wistron and E4 Computing, which showcased their newly announced OCP-form factor POWER8 server. Featuring two POWER8 processors, four NVIDIA Tesla P100 GPUs with the NVLink interconnect, and liquid cooling, the new platform represents an ideal OCP-compliant HPC system.

Talik also reported IBM, Xilinx, and Alpha Data showed their line ups of several FPGA adaptors designed for both POWER8 and POWER9. Featuring PCIe Gen3, CAPI1.0 for POWER8 and PCIe Gen4, CAPI2.0 and 25G/s CAPI3.0 for POWER9 these new FPGAs bring acceleration to a whole new level. OpenPOWER member engineers were on-hand to provide information regarding the CAPI SNAP developer and programming framework as well as OpenCAPI.

Not to be left out, Talik reported that IBM showcased products it previously tested and demonstrated: POWER8-based OCP and OpenPOWER Barreleye servers running IBM’s Spectrum Scale software, a full-featured global parallel file system with roots in HPC and now widely adopted in commercial enterprises across all industries for data management at petabyte scale.  Guess compute platform isn’t quite the dirty phrase IBM has been implying for months.

DancingDinosaur is Alan Radding, a veteran information technology analyst, writer, and ghost-writer. Please follow DancingDinosaur on Twitter, @mainframeblog. See more of his IT writing at technologywriter.com and here.

 

z System-Power-Storage Still Live at IBM

January 5, 2017

A mid-December briefing by Tom Rosamilia, SVP, IBM Systems, reassured some that IBM wasn’t putting its systems and platforms on the backburner after racking up financial quarterly losses for years. Expect new IBM systems in 2017. A few days later IBM announced that Japan-based APLUS Co., Ltd., which operates credit card and settlement service businesses, selected IBM LinuxONE as its mission-critical system for credit card payment processing. Hooray!

linuxone-emperor-2

LinuxONE’s security and industry-leading performance will ensure APLUS achieves its operational objectives as online commerce heats up and companies rely on cloud applications to draw and retain customers. Especially in Japan, where online and mobile shopping has become increasingly popular, the use of credit cards has grown, with more than 66 percent of consumers choosing that method for conducting online transactions. And with 80 percent enterprise hybrid cloud adoption predicted by 2017, APLUS is well positioned to connect cloud transactions leveraging LinuxONE. Throw in IBM’s expansion of blockchain capabilities and the APLUS move looks even smarter.

With the growth of international visitors spending money, IBM notes, and the emergence of FinTech firms in Japan have led to a diversification of payment methods the local financial industry struggles to respond. APLUS, which issues well-known credit cards such as T Card Plus, plans to offer leading-edge financial services by merging groups to achieve lean operations and improved productivity and efficiency. Choosing to update its credit card payment system with LinuxONE infrastructure, APLUS will benefit from an advanced IT environment to support its business growth by helping provide near-constant uptime. In addition to updating its server architecture, APLUS has deployed IBM storage to manage mission-critical data, the IBM DS8880 mainframe-attached storage that delivers integration with IBM z Systems and LinuxONE environments.

LinuxONE, however, was one part of the IBM Systems story Rosamilia set out to tell.  There also is the z13s, for encrypted hybrid clouds and the z/OS platform for Apache Spark data analytics and even more secure cloud services via blockchain on LinuxONE, by way of Bluemix or on premises.

z/OS will get attention in 2017 too. “z/OS is the best damn OLTP system in the world,” declared Rosamilia. He went on to imply that enhancements and upgrades to key z systems were coming in 2017, especially CICS, IMS, and a new release of DB2. Watch for new announcements coming soon as IBM tries to push z platform performance and capacity for z/OS and OLTP.

Rosamilia also talked up the POWER story. Specifically, Google and Rackspace have been developing OpenPOWER systems for the Open Compute Project.  New POWER LC servers running POWER8 and the NVIDIA NVLink accelerator, more innovations through the OpenCAPI Consortium, and the team of IBM and Nvidia to deliver PowerAI, part of IBM’s cognitive efforts.

As much as Rosamilia may have wanted to talk about platforms and systems IBM continues to avoid using terms like systems and platforms. So Rosamilia’s real intent was to discuss z and Power in conjunction with IBM’s strategic initiatives.  Remember these: cloud, big data, mobile, analytics. Lately, it seems, those initiatives have been culled down to cloud, hybrid cloud, and cognitive systems.

IBM’s current message is that IT innovation no longer comes from just the processor. Instead, it comes through scaling performance by workload and sustaining leadership through ecosystem partnerships.  We’ve already seen some of the fruits of that innovation through the Power community. Would be nice to see some of that coming to the z too, maybe through the open mainframe project. But that isn’t about z/0S. Any boost in CICS, DB2, and IMS will have to come from the core z team. The open mainframe project is about Linux on z.

The first glimpse we had of this came last spring in a system dubbed Minsky, which was described back then by commentator Timothy Prickett Morgan. With the Minsky machine, IBM is using NVLink ports on the updated Power8 CPU, which was shown in April at the OpenPower Summit and is making its debut in systems actually manufactured by ODM Wistron and rebadged, sold, and supported by IBM. The NVLink ports are bundled up in a quad to deliver 80 GB/sec bandwidth between a pair of GPUs and between each GPU and the updated Power8 CPU.

The IBM version, Morgan describes, aims to create a very brawny node with very tight coupling of GPUs and CPUs so they can better share memory, have fewer overall GPUs, and more bandwidth between the compute elements. IBM is aiming Minsky at HPC workloads, according to Morgan, but there is no reason it cannot be used for deep learning or even accelerated databases.

Is this where today’s z data center managers want to go?  No one is likely to spurn more performance, especially if it is accompanied with a price/performance improvement.  Whether rank-and-file z data centers are queueing up for AI or cognitive workloads will have to be seen. The sheer volume and scale of expected activity, however, will require some form of automated intelligent assist.

DancingDinosaur is Alan Radding, a veteran information technology analyst, writer, and ghost-writer. Please follow DancingDinosaur on Twitter, @mainframeblog. See more of his IT writing at technologywriter.com and here

Revamped IBM Power Systems LC Takes on x86

September 9, 2016

To hear IBM, its revamped and refreshed Power Systems LC lineup will undermine x86 (Intel), HPE, Dell/EMC, and any other purveyor of x86-based systems. Backed by accelerators provided by OpenPower community members, IBM appears ready extend the x86 battle to on premises, in the cloud, and the hybrid cloud. It promises to deliver better performance at lower cost for all the hot workloads too: artificial intelligence, deep learning, high performance data analytics, and compute-heavy workloads.

ibm-power-systems-s821lc

Two POWER8 processors, 1U config, priced 30% less than an x86 server

Almost a year ago, Oct. 2015, DancingDinosaur covered IBM previous Power Systems LC announcement here. The LC designation stands for Linux Community, and the company is tapping accelerators and more from the OpenPower community, just as it did with its recent announcement of POWER9 expected in 2017, here.

The new Power LC systems feature a set of community delivered technologies IBM has dubbed POWERAccel, a family of I/O technologies designed to deliver composable system performance enabled by accelerators. For GPU acceleration the NVDIA NVLink delivers nearly 5x better integration between POWER processors and the NVIDIA GPUs.  For FPGA acceleration IBM tapped its own CAPI architecture to integrate accelerators that run natively as part of the application.

This week’s Power Systems LC announcement features three new machines:

  • S821LC (pictured above)—includes 2 POWER8 sockets in a 1U enclosure and intended for environments requiring dense computing.
  • S822LC—brings 2 POWER8 sockets for big data workloads and adds big data acceleration through CAPI and GPUs.
  • S822LC—intended for high performance computing, it incorporates the new POWER8 processor with the NVDIA NVLink to deliver 2.8x the bandwidth to GPU accelerators and up to 4 integrated NVIDIA Pascal GPUs.

POWER8 with NVLink delivers 2.8 x the bandwidth compared to a PCle data pipe. According to figures provided by IBM comparing the price-performance of the Power S822LC for HPC (20-core, 256 GB, 4x Pascal) with a Dell C4130 (20-core, 256 GB 4xK80) and measured by total queries per hour (gph) the Power System delivered 2.1x better price-performance.  The Power Systems server cost more ($66,612) vs. the Dell ($57,615) but the Power System delivered 444 qph vs. Dell’s 185 qph.

The story plays out similarly for big data workloads running MongoDB on the IBM Power S8221LC for big data (20-core, 128 GB) vs. an HP DL380 (20-core, 128 GB). Here the system cost (server, OS, MongoDB annual subscription) came to $24,870 for IBM Power and $29,915 for HP.  Power provided 40% more performance at a 31% lower hardware/maintenance cost.

When it comes to the cloud the new IBM Power Systems LC offerings get even more interesting from a buyer’s standpoint. IBM declared the cloud a strategic imperative about 2 years ago and needs to demonstrate adoption that can rival the current cloud leaders; AWS, Google, and Microsoft (Azure). To that end IBM has started to tack on free cloud usage.

For example, during the industry analyst launch briefing IBM declared: Modernize your Power infrastructure for the Cloud, get access to IBM Cloud for free and cut your current operating costs by 50%. Whether you’re talking on-premises cloud or hybrid infrastructure the freebies just come. The free built-in cloud deployment service options include:

  • Cloud Provisioning and Automation
  • Infrastructure as a Service
  • Cloud Capacity Pools across Data Centers
  • Hybrid Cloud with BlueMix
  • Automation for DevOps
  • Database as a Service

These cover both on-premises, where you can transform your traditional infrastructure with automation, self-service, and elastic consumption models or a hybrid infrastructure where you can securely extend to Public Cloud with rapid access to compute services and API integration. Other freebies include open source automation, installation and configuration recipes, cross data center inventory, performance monitoring via the IBM Cloud, optional DR as a service for Power, and free access and capacity flexibility with SolfLayer (12 month starter pack).

Will the new LC line and its various cloud freebies get the low cost x86 monkey off IBM’s back? That’s the hope in Armonk. The new LC servers can be acquired at a lower price and can deliver 80% more performance per dollar spent over x86-based systems, according to IBM. This efficiency enables businesses and cloud service providers to lower costs and combat data center sprawl.

DancingDinosaur has developed TCO and ROI analyses comparing mainframe and Power systems to x86 for a decade, maybe more.  A few managers get it, but most, or their staff, have embedded bias and will never accept non-x86 machines. To them, any x86 system always is cheaper regardless of the specs and the math. Not sure even free will change their minds.

The new Power Systems LC lineup is price-advantaged over comparatively configured Intel x86-based servers, costing 30% less in some configurations.  Online LC pricing begins at $5999. Additional models with smaller configurations sport lower pricing through IBM Business Partners. All but the HPC machine are available immediately. The HPC machine will ship Sept. 26.

DancingDinosaur is Alan Radding, a veteran information technology analyst and writer. Please follow DancingDinosaur on Twitter, @mainframeblog. See more of his IT writing at technologywriter.com and here.

 

Oracle Aims at Intel and IBM POWER

July 8, 2016

In late June Oracle announced the SPARC S7 processor, a new 20nm, 4.27 GHz, 8-core/64-thread SPARC processor targeted for scale-out Cloud workloads that usually go to Intel x86 servers. These are among the same workloads IBM is aiming for with POWER8, POWER9, and eventually POWER10, as reported by DancingDinosaur just a couple of weeks ago.

oracle roadmap trajectory

Oracle 5-year SPARC trajectory (does not include newly announced S series).

According to Oracle, the latest additions to the SPARC platform are built on the new 4.27 GHz, 8-core/64-thread SPARC S7 microprocessor with what Oracle calls Software-in-Silicon features such as Silicon Secured Memory and Data Analytics Accelerators, which enable organizations to run applications of all sizes on the SPARC platform at commodity price points. All existing commercial and custom applications will also run on the new SPARC enterprise cloud services and solutions unchanged while experiencing improvements in security, efficiency, and simplicity.

By comparison, the IBM POWER platform includes with the POWER8, which is delivered as a 12-core, 22nm processor. The POWER9, expected in 2017, will be delivered as 14nm processor with 24 cores and CAPI and NVlink accelerators, which ensure delivery of more performance with greater energy efficiency.  By 2018, the IBM roadmap shows POWER8/9 as a 10nm, maybe even a 7nm, processor, based on the existing micro-architecture. And an even beefier POWER10 is expected to arrive around 2020.

At the heart of the Oracle’s new scale-out, commodity-priced server, the S7. According to Oracle, the SPARC S7 delivers balanced compute performance with 8 cores per processor, integrated on-chip DDR4 memory interfaces, a PCIe controller, and coherency links. The cores in the SPARC S7 are optimized for running key enterprise software, including Java applications and database. The SPARC S7–based servers use very high levels of integration that increase bandwidth, reduce latencies, simplify board design, reduce the number of components, and increase reliability, according to Oracle. All this promises an increase in system efficiency with a corresponding improvement in the economics of deploying a scale-out infrastructure when compared to other vendor solutions.

Oracle’s SPARC S7 processor, based on Oracle enterprise class M7 servers, is optimized for horizontally scalable systems with all the key functionality included in the microprocessor chip. Its Software-in-Silicon capabilities, introduced with the SPARC M7 processor, are also available in the SPARC S7 processor to enable improved data protection, cryptographic acceleration, and analytics performance. These features include Security-in-Silicon, which provides Silicon Secured Memory and cryptographic acceleration, and Data Analytics Accelerator (DAX) units, which provide In-memory query acceleration and in-line decompression

SPARC S7 processor–based servers include single- and dual-processor systems that are complementary to the existing mid-range and high-end systems based on Oracle’s SPARC M7 processor. SPARC S7 processor–based servers include two rack-mountable models. The SPARC S7-2 server uses a compact 1U chassis, and the SPARC S7-2L server is implemented in a larger, more expandable 2U chassis. Uniformity of management interfaces and the adoption of standards also should help reduce administrative costs, while the chassis design provides density, efficiency, and economy as increasingly demanded by modern data centers. Published reports put the cost of the new Oracle systems at just above $11,000 with a single processor, 64GB of memory and two 600GB disk drives, and up to about $50,000 with two processors and a terabyte of memory.

DancingDinosaur doesn’t really have enough data to compare the new Oracle system with the new POWER8 and upcoming POWER9 systems. Neither Oracle nor IBM have provided sufficient details. Oracle doesn’t even offer a roadmap at this point, which might tell you something.

What we do know about the POWER machines is this: POWER9 promises a wealth of improvements in speeds and feeds. Although intended to serve the traditional Power Server market, it also is expanding its analytics capabilities and is being optimized for new deployment models like hyperscale, cloud, and technical computing through scale-out deployment. Available for either clustered or multiple formats, it will feature a shorter pipeline, improved branch execution, and low latency on the die cache as well as PCI gen 4.

According to IBM, you can expect a 3x bandwidth improvement with POWER9 over POWER8 and a 33% speed increase. POWER9 also will continue to speed hardware acceleration and support next gen NVlink, improved coherency, enhance CAPI, and introduce a 25 GPS high speed link. Although the 2-socket chip will remain, IBM suggests larger socket counts are coming. It will need that to compete with Intel.

At least IBM showed its POWER roadmap. There is no comparable information from Oracle. At best, DancingDinosaur was able to dig up the following sketchy details for 2017-2019: Next Gen Core, 2017 Software-in-Silicon V1, Scale Out fully integrated Software-in-Silicon V1 or 2; 2018- 2019 Core Enhancements, Increased Cache, Increased Bandwidth, Software-in-Silicon V3.

Both Oracle and IBM have made it clear neither really wants to compete in the low cost, scale out server market. However, as both companies’ large clients turn to scale out, hyperscale Intel-based systems they have no choice but to follow the money. With the OpenPOWER Foundation growing and driving innovation, mainly in the form of accelerators, IBM POWER may have an advantage driving a very competitive price/performance story against Intel. With the exception of Fujitsu as an ally of sorts, Oracle has no comparable ecosystem as far as DancingDinosaur can tell.

DancingDinosaur is Alan Radding, a veteran information technology analyst and writer. Please follow DancingDinosaur on Twitter, @mainframeblog. See more of his IT writing at technologywriter.com and here.

 

IBM Fires a Shot at Intel with its Latest POWER Roadmap

June 17, 2016

In case you worry that IBM will abandon hardware in the pursuit of its strategic initiatives focusing on cloud, mobile, analytics and more; well, stop worrying. With the announcement of its POWER Roadmap at the OpenPOWER Summit earlier this spring, it appears POWER will be around for years to come. But IBM is not abandoning the strategic initiatives either; the new Roadmap promises to support new types of workloads, such as real time analytics, Linux, hyperscale data centers, and more along with support for the current POWER workloads.

power9b

Pictured above: POWER9 Architecture, courtesy of IBM

Specifically, IBM is offering a denser roadmap, not tied to technology and not even tied solely to IBM. It draws on innovations from a handful of the members of the Open POWER Foundation as well as support from Google. The new roadmap also signals IBM’s intention to make a serious run at Intel’s near monopoly on enterprise server processors by offering comparable or better price, performance, and features.

Google, for example, reports porting many of its popular web services to run on Power systems; its toolchain has been updated to output code for x86, ARM, or Power architectures with the flip of a configuration flag. Google, which strives to be everything to everybody, now has a highly viable alternative to Intel in terms of performance and price with POWER. At the OpenPOWER Summit early in the spring, Google made it clear it plans to build scale-out server solutions based on OpenPower.

Don’t even think, however, that Google is abandoning Intel. The majority of its systems are Intel-oriented. Still, POWER and the OpenPOWER community will provide a directly competitive processing alternative.  To underscore the situation Google and Rackspace announced they were working together on Power9 server blueprints for the Open Compute Project, designs that reportedly are compatible with the 48V Open Compute racks Google and Facebook, another hyperscale data center, already are working on.

Google represents another proof point that OpenPOWER is ready for hyperscale data centers. DancingDinosaur, however, really is interested most in what is coming from OpenPOWER that is new and sexy for enterprise data centers, since most DancingDinosaur readers are focused on the enterprise data center. Of course, they still need ever better performance and scalability too. In that regard OpenPOWER has much for them in the works.

For starters, POWER8 is currently delivered as a 12-core, 22nm processor. POWER9, expected in 2017, will be delivered as 14nm processor with 24 cores and CAPI and NVlink accelerators. That is sure to deliver more performance with greater energy efficiency.  By 2018, the IBM roadmap shows POWER8/9 as a 10nm, maybe even 7nm, processor, based on the existing micro-architecture.

The real POWER future, arriving around 2020, will feature a new micro-architecture, sport new features and functions, and bring new technology. Expect much, if not almost all, of the new functions to come from various OpenPOWER Foundation partners,

POWER9, only a year or so out, promises a wealth of improvements in speeds and feeds. Although intended to serve the traditional Power Server market, it also is expanding its analytics capabilities and bringing new deployment models for hyperscale, cloud, and technical computing through scale out deployment. This will include deployment in both clustered or multiple formats. It will feature a shorter pipeline, improved branch execution, and low latency on the die cache as well as PCI gen 4.

Expect a 3x bandwidth improvement with POWER9 over POWER8 and a 33% speed increase. POWER9 also will continue to speed hardware acceleration and support next gen NVlink, improved coherency, enhance CAPI, and introduce a 25 GPS high speed link. Although the 2-socket chip will remain, IBM suggests larger socket counts are coming. It will need that to compete with Intel.

As a data center manager, will a POWER9 machine change your data center dynamics?  Maybe, you decide: a dual-socket Power9 server with 32 DDR4 memory slots, two NVlink slots, three PCIe gen-4 x16 slots, and a total 44 core count. That’s a lot of computing power in one rack.

Now IBM just has to crank out similar advances for the next z System (a z14 maybe?) through the Open Mainframe Project.

DancingDinosaur is Alan Radding, a veteran information technology analyst and writer. Please follow DancingDinosaur on Twitter, @mainframeblog. See more of his IT writing at technologywriter.com and here.

 

IBM Advances SSD with Phase-Change Memory Breakthrough

May 20, 2016

Facing an incessant demand to speed data through computers the latest IBM storage memory advance, announced earlier this week, will ratchet up the speed another notch or two. Scientists at IBM Research have demonstrated storing 3 bits of data per cell using phase-change memory (PCM). Until now, PCM had been tried but had never caught on for a variety of reasons. By storing 3 bits per cell, IBM can boost PCM capacity and speed and lower the cost.

TLCPCMSmall (1)

IBM multi-bit PCM chip connected to a standard integrated circuit board.

Pictured above, the chip consists of a 2 × 2 Mcell array with a 4- bank interleaved architecture, IBM explained. The memory array size is 2 × 1000 μm × 800 μm. The PCM cells are based on doped-chalcogenide alloy and were integrated into the prototype chip serving as a characterization vehicle in 90 nm CMOS baseline technology.

Although PCM has been around for some years only with this latest advance is it attracting the industry’s attention as a potential universal memory technology based on its combination of read/write speed, endurance, non-volatility, and density. Specifically, PCM doesn’t lose data when powered off, unlike DRAM, and the technology can endure at least 10 million write cycles, compared to an average flash USB stick, which tops out at 3,000 write cycles.  Primary use cases will be capturing massive volumes of data expected from mobile devices and the Internet of Things.

PCM, in effect, adds another tier to the storage/memory hierarchy, coming in between DRAM and Flash at the upper levels of the storage performance pyramid. The IBM researchers envision both standalone PCM and hybrid applications, which combine PCM and flash storage together. For example, PCM can act as an extremely fast cache by storing a mobile phone’s operating system and enabling it to launch in seconds. For enterprise data centers, IBM envisions entire databases could be stored in PCM for blazing fast query processing of time-critical online applications, such as financial transactions.

As reported by CNET, PCM fits neatly between DRAM and flash. DRAM is 5-10x faster at retrieving data than PCM, while PCM is about 70x faster than flash. IBM reportedly expects PCM to be cheaper than DRAM, eventually becoming as cheap as flash (or course flash keeps getting cheaper too). PCM’s ability to hold three bits of data rather than 2 bits, PCM’s previous best, enables packing more data into a chip, which lowers the cost of PCM storage and boosts its competitive position against technologies like Flash and DRAM.

Phase change memory is the first instantiation of a universal memory with properties of both DRAM and flash, thus answering one of the grand challenges of our industry,” wrote Haris Pozidis, key researcher and manager of non-volatile memory research at IBM Research –in the published announcement. “Reaching 3 bits per cell is a significant milestone because at this density the cost of PCM will be significantly less than DRAM and closer to flash.”

IBM explains how PCM works: PCM materials exhibit two stable states, the amorphous (without a clearly defined structure) and crystalline (with structure) phases, of low and high electrical conductivity, respectively. In digital systems, data is stored as a 0 or a 1. To store a 0 or a 1 on a PCM cell, a high or medium electrical current is applied to the material. A 0 can be programmed to be written in the amorphous phase or a 1 in the crystalline phase, or vice versa. Then to read the bit back, a low voltage is applied.

To achieve multi-bit storage IBM scientists have developed two innovative enabling technologies: 1) a set of drift-immune cell-state metrics and 2) drift-tolerant coding and detection schemes. These new cell-state metrics measure a physical property of the PCM cell that remains stable over time, and are thus insensitive to drift, which affects the stability of the cell’s electrical conductivity with time. The other measures provide additional robustness of the stored data. As a result, the cell state can be read reliably over long time periods after the memory is programmed, thus offering non-volatility.

Combined these advancements address the key challenges of multi-bit PCM—drift, variability, temperature sensitivity and endurance cycling, according to IBM. From there, the experimental multi-bit PCM chip used by IBM scientists is connected to a standard integrated circuit board

Expect to see PCM first in Power Systems. At the 2016 OpenPOWER Summit in San Jose, CA, last month, IBM scientists demonstrated PCM attached to POWER8-based servers (made by IBM and TYAN® Computer Corp.) via the CAPI (Coherent Accelerator Processor Interface) protocol, which speeds the data to storage or memory. This technology leverages the low latency and small access granularity of PCM, the efficiency of the OpenPOWER architecture, and the efficiency of the CAPI protocol, an example of the OpenPower Foundation in action. Pozidis suggested PCM could be ready by 2017; maybe but don’t bet on it. IBM still needs to line up chip makers to produce it in commercial quantities among other things.

DancingDinosaur is Alan Radding, a veteran information technology analyst and writer. Please follow DancingDinosaur on Twitter, @mainframeblog. See more of his IT writing at technologywriter.com and here.

 

 

IBM’s Strategic Initiatives Gain New All-Flash Storage

May 6, 2016

Flash storage must be the latest rage among enterprise storage vendors.  Last week IBM introduced three new all-flash storage arrays, driving down latency and price/gigabyte to unheard of levels (minimum latency of 250μs, all-flash storage as low as $1.50 per gigabyte). Earlier this week EMC announced new all-flash arrays for its Unity series at prices under $18,000 (under $10,000 for hybrid arrays.) Flash storage has long beaten hard disk in terms of cost per IOPS, but now it is rivaling hard disk in terms of cost/gigabyte.

IBM_Flash_2015_1259-C-no_shadow_A9000GlamShot2

IBM A9000 All-Flash Array

OK, it looks a little—uh—boxy to say the least. But the new FlashSystem A9000 is packed with storage goodies. It comes fully configured, which helps drive down the cost of implementing an all-flash environment. Its sister, the FlashSystem A9000R, brings a grid architecture that provides for easy scaling up to the petabyte range. Both FlashSystems incorporate data reduction features, including pattern removal, deduplication and real-time compression, as well as IBM FlashCore technology to deliver consistent low latency performance. As noted above, they are priced as low as $1.50 per gigabyte.

Driving IBM’s latest interest in flash storage are its strategic initiatives, start with cloud computing. Consumers today, notes IBM, are demanding cloud-based applications that are fast, easy, and intelligent. That means minimal latency. Cloud users are demanding sub-second response times, especially when accessing critical data. They also are demanding cloud providers deliver a unique, personalized, and positive customer experience.

To deliver this, IBM is turning to hardware innovation, specifically its MicroLatency technology, to transfers data within the flash array instead of adding another layer of software. MicroLatency technology inserts FPGAs (hardware) that connects and communicates directly with the flash and RAID controllers, eliminating the latency of software and even firmware. Instead, the FlashSystems lets hardware talk directly with hardware.

In addition, IBM is packing the new FlashSystem arrays with features designed to solve cloud requirements such as quality-of-service (QoS) to prevent the noisy neighbor problems with application performance. The new arrays also feature secure multi-tenancy, thresholding, and easy-to-deploy grid scale-out capabilities.

The z System platform is not being ignored in all of this. IBM is including a new DS model, the all-flash IBM DS 8888 optimized for enterprise-class servers: With the all-flash IBM DS8888, customer databases and data-intensive applications are accelerated, resulting in improved business performance and customer satisfaction.

Specifically, the DS888 brings faster decision making and improve customer serviceability, with 4x performance over previous generations and accelerated response time for mission critical applications. The flash storage delivers up to 2.5 million IOPS, the result of having been built on the Power8 processor. It also enables organizations to streamline operations through the performance of an all flash architected solution aligned to provide the deepest integration with System z environments. For instance, IBM promises the most robust FICON connectivity through an architecture optimized for mainframe’s 4K cache segments.

In addition, the DS8888 promises 24×7 access to data and applications through superior business continuity on high demand transaction processing workloads while delivering top operations performance through its all flash architecture. It goes beyond the usual high end 5-nines availability to deliver 6-nines availability, which translates into a mere 2.59 seconds of downtime per month.  Other availability features include flexible replication (IBM FlashCopy, Metro Mirror, Global Mirror, Metro/Global Mirror, Global Copy & Multiple Target Peer-to-Peer Remote Copy). In the early years of flash reliability and availability were a concern.  With the DS8888 and 6-nines availability it isn’t any more.

Finally, it comes with a smorgasbord of security and efficiency goodies, including self-encrypted flash drives, key interoperability management protocol, syslog protocol, an intuitive GUI (IBM has learned a few tricks from Apple), innovative storage software licensing, RESTful and OpenStack APIs to connect workloads between private and public clouds, and thin provisioning for maximum utilization and reclamation of capacity from deleted data.

All-flash solutions announced last week complement IBM’s existing all-flash portfolio including FlashSystem 900 and V9000 that also leverage IBM’s FlashCore technology. IBM’s midrange all-flash solutions consist of all-flash versions of IBM’s Storwize family, which offers the performance needed for real-time insights from business data combined with advanced management functions. IBM’s Big Data all-flash solution delivers high-density multi-petabyte scale and a low-cost flash option ideal for industries such as media, genomics, and life sciences.

DancingDinosaur used to be hired to write papers around the enterprise cost-performance tradeoffs between hard disk and SSD/flash. No matter how expensive flash was at whatever point, the cost per IOPS always favored flash and cost per gigabytes always favored hard disk. That’s no longer an analysis worth even making today.

DancingDinosaur is Alan Radding, a veteran information technology analyst and writer. Please follow DancingDinosaur on Twitter, @mainframeblog. See more of his IT writing at technologywriter.com and here.

 

 

Ubuntu Linux (beta) for the z System is Available Now

April 8, 2016

As recently as February, DancingDinosaur has been lauding IBM’s bolstering of the z System for Linux and support for the latest styles of app dev. As part of that it expected Ubuntu Linux for z by the summer. It arrived early.  You can download it for LinuxONE and the z now, hereubuntu-logo-300x225

Of course, the z has run Linux for over a decade. That was a customized version that required a couple of extra steps, mainly recompiling, if x86 Linux apps were to run seamlessly. This time Canonical and the Ubuntu community have committed to work with IBM to ensure that Ubuntu works seamlessly with IBM LinuxONE, z Systems, and Power Systems. The goal is to enable IBM’s enterprise platforms to play nicely with the latest app dev goodies, including NFV, containers, KVM, OpenStack, big data analytics, DevOps, and even IoT. To that end, all three parties (Canonical, the Ubuntu community, and IBM) commit to provide reference architectures, supported solutions, and cloud offerings, now and in the future.

Ubuntu is emerging as the platform of choice for organizations running scale-out, next-generation workloads in the cloud. According to Canonical, Ubuntu dominates public cloud guest volume and production OpenStack deployments with up to 70% market share. Global brands running Ubuntu at scale in the cloud include AT&T, Walmart, Deutsche Telecom, Bloomberg, Cisco and others.

The z and LinuxONE machines play right into this. They can support thousands of Linux images with no-fail high availability, security, and performance. When POWER 9 processors come to market it gets even better. At a recent OpenPOWER gathering the POWER 9 generated tremendous buzz with Google discussing its intentions of building a new data center server  based on an open POWER9 design that conforms to Facebook’s Open Compute Project server.

These systems will be aimed initially at hyperscale data centers. OpenPOWER processors combined with acceleration technology have the potential to fundamentally change server and data center design today and into the future.  OpenPOWER provides a great platform for the speed and flexibility needs of hyperscale operators as they demand ever-increasing levels of scalability.

According to Aaron Sullivan, Open Compute Project Incubation Committee Member and Distinguished Engineer at Rackspace. “OpenPOWER provides a great platform for the speed and flexibility needs of hyperscale operators as they demand ever-increasing levels of scalability.” This is true today and with POWER9, a reportedly 14nm processor coming around 2017, it will be even more so then. This particular roadmap looks out to 2020 when POWER10, a 10nm processor, is expected with the task of delivering extreme analytics optimization.

But for now, what is available for the z isn’t exactly chopped liver. Ubuntu is delivering scale-out capabilities for the latest development approaches to run on the z and LinuxONE. As Canonical promises: Ubuntu offers the best of open source for IBM’s enterprise customers along with unprecedented performance, security and resiliency. The latest Ubuntu version, Ubuntu 16.04 LTS, is in beta and available to all IBM LinuxOne and z Systems customers. See the link above. Currently SUSE and Red Hat are the leading Linux distributions among z data centers. SUSE also just announced a new distro of openSUSE Linux for the z to be called openSUSE Factory.

Also this week the OpenPOWER Foundation held its annual meeting where it introduced technology to boost data center infrastructures with more choices, essentially allowing increased data workloads and analytics to drive better business results. Am hoping that the Open Mainframe Project will emulate the Open POWER group and in a year or two by starting to introducing technology to boost mainframe computing along the same lines.

For instance OpenPOWER introduced more than 10 new OpenPOWER servers, offering expanded services for high performance computing and server virtualization. Or this: IBM, in collaboration with NVIDIA and Wistron, revealed plans to release its second-generation OpenPOWER high performance computing server, which includes support for the NVIDIA Tesla Accelerated Computing platform. The server will leverage POWER8 processors connected directly to the new NVIDIA Tesla P100 GPU accelerators via the NVIDIA NVLink, a high-speed interconnect technology.

In the same batch of announcements TYAN announced its GT75-BP012, a 1U, POWER8-based server solution with the ppc64 architecture. The ppc64 architecture is optimized for 64-bit big-endian PowerPC and Power Architecture processors.  Also of interest to DancingDinosaur readers may be the variation of the ppc64 that enables a pure little-endian mode with the POWER8 to enable the porting of x86 Linux-based software with minimal effort. BTW, the OpenPOWER-based platform, reportedly, offers exceptional capability for in-memory computing in a 1U implementation, part of the overall trend toward smaller, denser, and more efficient systems. The latest TYAN offerings will only drive more of it.

DancingDinosaur is Alan Radding, a veteran information technology analyst and writer. Please follow DancingDinosaur on Twitter, @mainframeblog. See more of his IT writing at technologywriter.com and here.


%d bloggers like this: