Posts Tagged ‘cognitive workloads’

Get a Next-Gen Datacenter with IBM-Nutanix POWER8 System

July 14, 2017

First announced by IBM on May 16 here, this solution, driven by client demand for a simplified hyperconverged—combined server, network, storage, hardware, software—infrastructure, is designed for data-intensive enterprise workloads.  Aimed for companies increasingly looking for the ease of deployment, use, and management that hyperconverged solutions promise. It is being offered as an integrated hardware and software offering in order to deliver on that expectation.

Music made with IBM servers, storage, and infrastructure

IBM’s new POWER8 hyperconverged solutions enable a public cloud-like experience through on-premises infrastructure with top virtualization and automation capabilities combined with Nutanix’s public and on-premises cloud capabilities. They provide a combination of reliable storage, fast networks, scalability and extremely powerful computing in modular, scalable, manageable building blocks that can be scaled simply by adding nodes when needed.

Over time, IBM suggests a roadmap of offerings that will roll out as more configurations are needed to satisfy client demand and as feature and function are brought into both the IBM Cognitive Systems portfolio and the Nutanix portfolio. Full integration is key to the value proposition of this offering so more roadmap options will be delivered as soon as feature function is delivered and integration testing can be completed.

Here are three immediate things you might do with these systems:

  1. Mission-critical workloads, such as databases, large data warehouses, web infrastructure, and mainstream enterprise apps
  2. Cloud native workloads, including full stack open source middleware, enterprise databases
    and containers
  3. Next generation cognitive workloads, including big data, machine learning, and AI

Note, however, the change in IBM’s pricing strategy. The products will be priced with the goal to remain neutral on total cost of acquisition (TCA) to comparable offerings on x86. In short, IBM promises to be competitive with comparable x86 systems in terms of TCA. This is a significant deviation from IBM’s traditional pricing, but as we have started to see already and will continue to see going forward IBM clearly is ready to play pricing flexibility to win the deals on products it wants to push.

IBM envisions the new hyperconverged systems to bring data-intensive enterprise workloads like EDB Postgres, MongoDB and WebSphere into a simple-to-manage, on-premises cloud environment. Running these complex workloads on IBM Hyperconverged Nutanix POWER8 system can help an enterprise quickly and easily deploy open source databases and web-serving applications in the data center without the complexity of setting up all of the underlying infrastructure plumbing and wrestling with hardware-software integration.

And maybe more to IBM’s ultimate aim, these operational data stores may become the foundational building blocks enterprises will use to build a data center capable of taking on cognitive workloads. These ever-advancing workloads in advanced analytics, machine learning and AI will require the enterprise to seamlessly tap into data already housed on premises. Soon expect IBM to bring new offerings to market through an entire family of hyperconverged systems that will be designed to simply and easily deploy and scale a cognitive cloud infrastructure environment.

Currently, IBM offers two systems: the IBM CS821 and IBM CS822. These servers are the industry’s first hyperconverged solutions that marry Nutanix’s one-click software simplicity and scalability with the proven performance of the IBM POWER architecture, which is designed specifically for data-intensive workloads. The IBM CS822 (the larger of the two offerings) sports 22 POWER8 processor cores. That’s 176 compute threads, with up to 512 GB of memory and 15.36 TB of flash storage in a compact server that meshes seamlessly with simple Nutanix Prism management.

This server runs Nutanix Acropolis with AHV and little endian Linux. If IBM honors its stated pricing policy promise, the cost should be competitive on the total cost of acquisition for comparable offerings on x86. DancingDinosaur is not a lawyer (to his mother’s disappointment), but it looks like there is considerable wiggle room in this promise. IBM Hyperconverged-Nutanix Systems will be released for general availability in Q3 2017. Specific timelines, models, and supported server configurations will be announced at the time of availability.

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.

 

Meet the POWER9 Chip Family

September 2, 2016

When you looked at a chip in the past you primarily were concerned with two things: the speed of the chip, usually expressed in GHz, and how much power it consumed. Today the IBM engineers preparing the newest POWER chip, the 14nm POWER9, are tweaking the chips for the different workloads it might run, such as cognitive or cloud, and different deployment options, such as scale-up or scale-out, and a host of other attributes.  EE Times described it in late August from the Hot Chips conference where it was publicly unveiled.

ibm power9 bandwidth

IBM POWER9 chip

IBM describes it as a chip family but maybe it’s best described as the product of an entire chip community, the Open POWER Foundation. Innovations include CAPI 2.0, New CAPI, Nvidia’s NVLink 2.0, PCle Gen4, and more. It spans a range of acceleration options from HSDC clusters to extreme virtualization capabilities for the cloud. POWER9 is not just about high speed transaction processing; IBM wants the chip to interpret and reason, ingest and analyze.

POWER has gone far beyond the POWER chips that enabled Watson to (barely) beat the human Jeopardy champions. Going forward, IBM is counting on POWER9 and Watson to excel at cognitive computing, a combination of high speed analytics and self-learning. POWER9 systems should not only be lightning fast but get smarter with each new transaction.

For z System shops, POWER9 offers a glimpse into the design thinking IBM might follow with the next mainframe, probably the z14 that will need comparable performance and flexibility. IBM already has set up the Open Mainframe Project, which hasn’t delivered much yet but is still young. It took the Open POWER group a couple of years to deliver meaningful innovations. Stay tuned.

The POWER9 chip is incredibly dense (below). You can deploy it as either a scale-up or scale-out architecture. You have a choice of two-socket servers with 8 DDR4 ports and another for multiple chips per server with buffered DIMMs.

power9 chip

IBM POWER9 silicon layout

IBM describes the POWER9 as a premier acceleration platform. That means it offers extreme processor/accelerator bandwidth and reduced latency; coherent memory and virtual addressing capability for all accelerators; and robust accelerated compute options through the OpenPOWER community.

It includes State-of-the-Art I/O and Acceleration Attachment Signaling:

  • PCIe Gen 4 x 48 lanes – 192 GB/s duplex bandwidth
  • 25G Link x 48 lanes – 300 GB/s duplex bandwidth

And robust accelerated compute options based on open standards, including:

  • On-Chip Acceleration—Gzip x1, 842 Compression x2, AES/SHA x2
  • CAPI 2.0—4x bandwidth of POWER8 using PCIe Gen 4
  • NVLink 2.0—next generation of GPU/CPU bandwidth and integration using 25G Link
  • New CAPI—high bandwidth, low latency and open interface using 25G Link

In scale-out mode it employs direct attached memory through 8 direct DDR4 ports, which deliver:

  • Up to 120 GB/s of sustained bandwidth
  • Low latency access
  • Commodity packaging form factor
  • Adaptive 64B / 128B reads

In scale-up mode it uses buffered memory through 8 buffered channels to provide:

  • Up to 230GB/s of sustained bandwidth
  • Extreme capacity – up to 8TB / socket
  • Superior RAS with chip kill and lane sparing
  • Compatible with POWER8 system memory
  • Agnostic interface for alternate memory innovations

POWER9 was publicly introduced at the Hot Chips conference last spring. Commentators writing in EE Times noted that POWER9 could become a break out chip, seeding new OEM and accelerator partners and rejuvenating IBM’s efforts against Intel in high-end servers. To achieve that kind of performance IBM deploys large chunks of memory—including a 120 Mbyte embedded DRAM in shared L3 cache while riding a 7 Tbit/second on-chip fabric. POWER9 should deliver as much as 2x the performance of the Power8 or more when the new chip arrives next year, according to Brian Thompto, a lead architect for the chip, in published reports.

As noted above, IBM will release four versions of POWER9. Two will use eight threads per core and 12 cores per chip geared for IBM’s Power virtualization environment; two will use four threads per core and 24 cores/chip targeting Linux. Both will come in two versions — one for two-socket servers with 8 DDR4 ports and another for multiple chips per server with buffered DIMMs.

The diversity of choices, according to Hot Chips observers, could help attract OEMs. IBM has been trying to encourage others to build POWER systems through its OpenPOWER group that now sports more than 200 members. So far, it’s gaining most interest from China where one partner plans to make its own POWER chips. The use of standard DDR4 DIMMs on some parts will lower barriers for OEMs by enabling commodity packaging and lower costs.

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: