Cisco shoves more GPUs in AI server for deep learning, still doesn't play Crysis

More power and faster interconnect


Cisco has beefed up its C480 AI/machine learning server, adding a faster GPU interconnect and more GPU slots while losing two CPU sockets.

Fairy in the woods

If you've got $1m+ to blow on AI, meet Pure, Nvidia's AIRI fairy: A hyperconverged beast

READ MORE

The C480 M5 is a 4U rackmount 4-socket Xeon modular server with up to 6TB of memory, six AMD S7150 x2 graphics chips or Nvidia Tesla P40 and Tesla P100 GPUs, up to 32 drives and 12 NVMe SSDs – 44 drives in all. It was revealed in June 2017.

Since then NetApp and Pure have introduced all-flash storage systems with more GPUs inside NVidia's DGX-1 GPU server system, interconnected by NVlink. Pure's can have four DGX-1s.

Cisco has announced an updated C480 ML M5 – "ML" of course standing for machine learning – with 2 x Xeon SP CPUs, up to 3TB of memory, 8 x Nvidia Tesla V100-32G, up to 24 x SAS/SATA HDD/SSD (to 182TB), plus 6 x NVMe drives, and up to 4 x x100G VIC (Virtual Interface Card), and NVlink. That's a total of 30 drives.

Cisco_UCS_C480_M5_ML

Cisco UCS C480 M5 ML

Cisco said it is a server for deep learning – a compute-intensive form of machine learning that uses neural networks and large data sets to train computers for complex tasks.

It has been working with Hortonworks to validate Hadoop 3.1 in a design where the UCS C480 ML is part of the big data cluster, storing data on the C480 ML disk drives, and supporting Docker containers running analytic workloads such as Apache Spark and Google TensorFlow that require both CPUs and GPUs. There is also a project with Cloudera.

More C480/Cisco AI ecosystem work involves Cisco contributing code to the Google Kubeflow open-source project, which integrates TensorFlow with Kubernetes, announcing its DevNet AI Developer Center for developers, operators, and data scientists, a DevNet Ecosystem Exchange, and working with Anaconda so data scientists and developers can collaborate on machine learning using languages such as Python.

Unlike NetApp and Pure, Cisco did not provide Resnet or Alexnet test results, saying that these don't reflect what it expects the reality of real-world AI/machine learning projects to be. It pointed out that the storage drives in the C480 ML are inside the enclosure and not connected across a storage network.

Cisco envisions the C480 being used in harness with C280 Hadoop systems, as well as other UCS servers, and said all UCS servers can be managed by the cloud-based Intersight systems management platform.

AI, AI, Pure: Nvidia cooks deep learning GPU server chips with NetApp

READ MORE

Cisco expects the enterprise market for A to grow significantly. It cited a Gartner report* saying "only 4 per cent of CIOs worldwide report that they have AI projects in production" at present. However, it also quoted a 2017 McKinsey AI report that predicts:

  • 75 per cent of developer teams will include AI functionality in one or more applications in 2018
  • 40 per cent of all digital transformation initiatives will be enabled by AI by 2019
  • 100 per cent of all effective IoT efforts will be supported by AI capabilities by 2019

AI and machine learning looks set to be a huge new market for server, GPU and flash storage array vendors for the next few years.

The UCS C480 ML M5 Rack Server will be available from Cisco partners in Q4 2018, along with a range of AI and ML support capabilities from Cisco Services that span analytics, deep learning and automation. ®

* Gartner Inc. Hype Cycle for Artificial Intelligence, 2018, Svetlana Sinclair and Kenneth Brant, July 24, 2018.

Similar topics

Broader topics


Other stories you might like

  • Cisco warns of security holes in its security appliances
    Bugs potentially useful for rogue insiders, admin account hijackers

    Cisco has alerted customers to another four vulnerabilities in its products, including a high-severity flaw in its email and web security appliances. 

    The networking giant has issued a patch for that bug, tracked as CVE-2022-20664. The flaw is present in the web management interface of Cisco's Secure Email and Web Manager and Email Security Appliance in both the virtual and hardware appliances. Some earlier versions of both products, we note, have reached end of life, and so the manufacturer won't release fixes; it instead told customers to migrate to a newer version and dump the old.

    This bug received a 7.7 out of 10 CVSS severity score, and Cisco noted that its security team is not aware of any in-the-wild exploitation, so far. That said, given the speed of reverse engineering, that day is likely to come. 

    Continue reading
  • Train once, run anywhere, almost: Qualcomm's drive to bring AI to its phone, PC chips
    Software toolkit offered to save developers time, effort, battery power

    Qualcomm knows that if it wants developers to build and optimize AI applications across its portfolio of silicon, the Snapdragon giant needs to make the experience simpler and, ideally, better than what its rivals have been cooking up in the software stack department.

    That's why on Wednesday the fabless chip designer introduced what it's calling the Qualcomm AI Stack, which aims to, among other things, let developers take AI models they've developed for one device type, let's say smartphones, and easily adapt them for another, like PCs. This stack is only for devices powered by Qualcomm's system-on-chips, be they in laptops, cellphones, car entertainment, or something else.

    While Qualcomm is best known for its mobile Arm-based Snapdragon chips that power many Android phones, the chip house is hoping to grow into other markets, such as personal computers, the Internet of Things, and automotive. This expansion means Qualcomm is competing with the likes of Apple, Intel, Nvidia, AMD, and others, on a much larger battlefield.

    Continue reading
  • Is computer vision the cure for school shootings? Likely not
    Gun-detecting AI outfits want to help while root causes need tackling

    Comment More than 250 mass shootings have occurred in the US so far this year, and AI advocates think they have the solution. Not gun control, but better tech, unsurprisingly.

    Machine-learning biz Kogniz announced on Tuesday it was adding a ready-to-deploy gun detection model to its computer-vision platform. The system, we're told, can detect guns seen by security cameras and send notifications to those at risk, notifying police, locking down buildings, and performing other security tasks. 

    In addition to spotting firearms, Kogniz uses its other computer-vision modules to notice unusual behavior, such as children sprinting down hallways or someone climbing in through a window, which could indicate an active shooter.

    Continue reading

Biting the hand that feeds IT © 1998–2022