Nvidia aims to unify AI, HPC computing in HGX-2 server platform

Nvidia is refining its pitch for data-center performance and efficiency with a new server platform, the HGX-2, designed to harness the power of 16 Tesla V100 Tensor Core GPUs to satisfy requirements for both AI and high-performance computing (HPC) workloads.Data-center server makers Lenovo, Supermicro, Wiwynn and QCT said they would ship HGX-2 systems by the end of the year. Some of the biggest customers for HGX-2 systems are likely to be hyperscale providers, so it's no surprise that Foxconn, Inventec, Quanta and Wistron are also expected to manufacture servers that use the new platform for cloud data centers.  The HGX-2 is built using two GPU baseboards that link the Tesla GPUs via NVSwitch interconnect fabric. The HGX-2 baseboards handle 8 processors each, for a total of 16 GPUs. The HGX-1, announced a year ago, handled only 8 GPUs.To read this article in full, please click here

Nvidia aims to unify AI, HPC computing in HGX-2 server platform

Nvidia is refining its pitch for data-center performance and efficiency with a new server platform, the HGX-2, designed to harness the power of 16 Tesla V100 Tensor Core GPUs to satisfy requirements for both AI and high-performance computing (HPC) workloads.Data-center server makers Lenovo, Supermicro, Wiwynn and QCT said they would ship HGX-2 systems by the end of the year. Some of the biggest customers for HGX-2 systems are likely to be hyperscale providers, so it's no surprise that Foxconn, Inventec, Quanta and Wistron are also expected to manufacture servers that use the new platform for cloud data centers.  The HGX-2 is built using two GPU baseboards that link the Tesla GPUs via NVSwitch interconnect fabric. The HGX-2 baseboards handle 8 processors each, for a total of 16 GPUs. The HGX-1, announced a year ago, handled only 8 GPUs.To read this article in full, please click here

Windows Updates and Ansible

Ansible-Get-Started-Windows

Welcome to the fourth installment of our Windows-centric Getting Started Series!

One of the duties of most IT departments is keeping systems up to date. In this post we’re taking a quick look at using Ansible to manage updates on your Windows nodes. Starting with a small example of six Windows machines, we’ll show an example of a play against those hosts. We’ll share the full example at the end.

Updates, Updates, Updates...

Managing Windows updates is something that can be understood and customized quickly with Ansible. Below is a small-scale example of running updates on hosts with some flexibility in what gets updated in the process. The example here is assuming a domain exists and the hosts are being passed domain credentials. If you’re looking to test this example, be sure to read Bianca’s earlier Getting Started post on connecting to a Windows host.

Because this example is running against exclusively Windows machines, the information needed to connect can be included in the inventory file:

[all:vars]
ansible_connection: winrm
ansible_user: administrator
ansible_password: This-Should-Be-a-Password!

For Example

The example hosts include three groups of servers, two in each group. There are terminal servers, application servers, and directory servers. For the purposes of Continue reading

BrandPost: The Next SDN Leap: Automation and Intent-Driven Networking

Without an agile, flexible, and secure network infrastructure, organizations are in danger of falling behind competitors. That’s why many organizations are seeking to transform their businesses with cloud computing and hybrid cloud environments that are more adaptive and flexible. Software-defined networks (SDN) and network functions virtualization (NFV) can ease that path, but it requires automation along with intelligence that understands and can even predict what users and organizations want and need to do.Digital transformation has quickly moved beyond hype to become one of the top business imperatives. “Digital transformation is forcing companies to be agile and move with speed, and the network needs to be equally agile and fast,” writes industry analyst Zeus Kerravala. “The separation of control and data planes enables control to be abstracted away from the device and centralized so a network administrator can issue a change that is propagated instantly across the entire network.”To read this article in full, please click here

IDG Contributor Network: Internet testing results: why fixing the internet middle mile is essential for SD-WAN performance

It’s no secret that the public Internet is a quagmire of latency and packet loss problems. No wonder, many of clients are reluctant to trust Internet-based SD-WANs with VoIP and business-critical applications. After all, how can an SD-WAN running over Internet provide a predictable user experience if the underlying transport is so unpredictable?To answer that question, SD-WAN Experts recently evaluated the performance and stability of long-distance Internet connections. Our goal: to determine the source of the Internet's performance problems by measuring variability and latency in the last and middle miles.What we found was by swapping out the Internet core for a managed middle mile makes an enormous difference. Case in point is Amazon. The latency and variation between our AWS workloads was significantly better across Amazon’s network than the public Internet (see figure). Why that’s the case and how we tested is explained below and in greater depth from this post on our site.To read this article in full, please click here

IDG Contributor Network: Internet testing results: why fixing the internet middle mile is essential for SD-WAN performance

It’s no secret that the public Internet is a quagmire of latency and packet loss problems. No wonder, many of clients are reluctant to trust Internet-based SD-WANs with VoIP and business-critical applications. After all, how can an SD-WAN running over Internet provide a predictable user experience if the underlying transport is so unpredictable?To answer that question, SD-WAN Experts recently evaluated the performance and stability of long-distance Internet connections. Our goal: to determine the source of the Internet's performance problems by measuring variability and latency in the last and middle miles.What we found was by swapping out the Internet core for a managed middle mile makes an enormous difference. Case in point is Amazon. The latency and variation between our AWS workloads was significantly better across Amazon’s network than the public Internet (see figure). Why that’s the case and how we tested is explained below and in greater depth from this post on our site.To read this article in full, please click here

The Virtual Cloud Network Demystified

Introduction

Welcome to Summer 2018!  It’s been nearly one month now since our CEO Pat Gelsinger announced the Virtual Cloud Network vision at Dell Technologies world in Las Vegas.  Essentially the reveal (in my personal opinion) was focused on raising awareness that VMware has now delivered to the market what many of you have heard for quite some time now as “the vision” for networking and security, whereas NSX has become an integral part of many various parts of your business:

Enter stage left, the Virtual Cloud Network.  VCN builds upon the fundamentals you’re already familiar with from NSX—these include (but are not limited to) integrated security, consistent connectivity, and inherit automation, but really focuses on tying together an end-to-end architecture that allows our customers to deliver applications and services everywhere.  Our customers have asked and we have listened… the demand for any infrastructure, any cloud, any transport, any device, and any application has drastically changed the landscape and technologies associated with building/architecting and having a modern enterprise network.

We’ve been quite busy over the past month with lots of interest coming from partners and customers wondering what this really means.  Well today the wait Continue reading

Worth Reading: Open Optical

Network operators are intrigued by the concept open optical systems, whereby they can assemble an optical network by piecing together parts from various vendors. They are encouraged by concepts like disaggregated transponder boxes for DCI applications, an Open ROADM MSA, and the Telecom Infra Project Voyager white box initiative. While there is undoubtedly increasing interest in the open model, the question is how far can and should this model be pushed. —Jonathan Homa @ECI

1 … 1,584 1,585 1,586 1,587 1,588 … 3,836