Friday, April 3, 2020

BrandPost: What is SR-IOV and Why Is it the Gold-Standard for GPU Sharing?

In virtualization, SR-IOV (Single Root Input/Output Virtualization) is a specification that allows the isolation of PCI Express resources between different users. It is already the standard used to share networking resources (NICs) and secure network traffic. Each resource has Virtual Functions (VF) associated and each VM (Virtual machine) can only access the physical resource via its own allocated VF.

The AMD MxGPU (GPU sharing technology) is the industry’s first SR-IOV-based GPU sharing technology designed for cloud and datacenter. So why did we choose SR-IOV?

To read this article in full, please click here


Thanks to AMD (see source)

Free ipSpace.net Content

Most of us are in some sort of lockdown (or quarantine or shelter-in-place or whatever it’s called) at the moment. Some have their hands full balancing work and homeschooling their kids (hang in there!), others are getting bored and looking for networking-related content (or you wouldn’t be reading this blog).

If you’re in the latter category you might want to browse some of the free ipSpace.net content: almost 3500 blog posts, dozens of articles, over a hundred podcast episodes, over 20 free webinars, and another 30+ webinars with sample videos that you can access with free subscription.

Need more? Standard subscription includes 260 hours of video content and if you go for Expert subscription and select the network automation course as part of the subscription, you’ll get another 60 hours of content plus hands-on exercises, support, access to Slack team… hopefully enough to last you way past the peak of the current pandemic.


Thanks to (see source)

Thursday, April 2, 2020

Colleges expand VPN capacity, conferencing to answer COVID-19

Colleges that moved from on-campus classrooms to remote learning due to COVID-19 had to quickly upgrade networks to support new VPN connections for remote access. Fortunately, many online-learning platforms rely on cloud-based applications that don’t put additional strain on campus networks.

For example, The College of the Holy Cross in Worcester, Mass., added extra VPN user licenses for students and staff now working from home. It also ramped up its VPN-server capacity, according to Dr. Ellen J. Keohane, the college’s CIO. “We’re definitely seeing higher demand on that.”

To read this article in full, please click here


Thanks to Keith Shaw (see source)

T-Mobile borrows $19bn to fund Sprint takeover

About $65bn of investor orders for deal underscore demand for investment-grade debt
Thanks to (see source)

Scheduling tasks on Linux using the at command

When you want commands or scripts to run at some particular time, you don’t have to sit with your fingers hovering over the keyboard waiting to press the enter key or even be at your desk at the right time. Instead, you can set your task to be run through the at command. In this post, we’ll look at how tasks are scheduled using at, how you can precisely select the time you want your process to run and how to view what’s been scheduled to run using at.

at vs cron

For those who’ve been scheduling tasks on Linux systems using cron, the at command is something like cron in that you can schedule tasks to run at a selected time, but cron is used for jobs that are run periodically – even if that means only once a year. Most cron jobs are set up to be run daily, weekly or monthly, though you control how often and when.

To read this article in full, please click here


Thanks to Sandra Henry-Stocker (see source)

FCC wants to add a new swath of bandwidth to Wi-Fi 6

A proposed FCC rule would allow Wi-Fi 6 devices to make unlicensed use of an additional range of wireless spectrum, which would more than quadruple the number of channels available to Wi-Fi routers.

While that might be good news for enterprises seeking higher density Wi-Fi deployments,  current license holders of the 6GHz spectrum are concerned about interference from unlicensed use.

The driving factor, as ever, is the bottomless demand for spectrum caused by the increasing use of wireless just about everywhere, and the FCC’s announcement cites projections from Cisco that say about 60% of worldwide data traffic will move across Wi-Fi links within the next two years. Using the full 6GHz spectrum – all 1,200MHz of it – is part of the Wi-Fi 6 (802.11ax) standard that can’t be put into use until it is freed up by the FCC. With that spectrum extension in place the standard is known as Wi-Fi 6E, and devices with new silicon would be needed to implement it. (Wi-Fi 6E products are expected to hit shelves by the 2020 holiday season, per PC World.)

To read this article in full, please click here


Thanks to Jon Gold (see source)

China Mobile picks Huawei and ZTE to build its 5G network

State-owned company’s $5.2bn contract will cover more than 200,000 base stations
Thanks to (see source)

MUST READ: Using BGP RPKI for a Safer Internet

As I explained in How Networks Really Work and Upcoming Internet Challenges webinars, routing security, and BGP security in particular remain one of the unsolved challenges we’ve been facing for decades (see also: what makes BGP a hot mess).

Fortunately, due to enormous efforts of a few persistent individuals BGP RPKI is getting traction (NTT just went all-in), and Flavio Luciani and Tiziano Tofoni decided to do their part creating an excellent in-depth document describing BGP RPKI theory and configuration on Cisco- and Juniper routers.

There are only two things you have to do:

Thank you, the Internet will be grateful.


Thanks to (see source)

Wednesday, April 1, 2020

Thousands of PCs break exaFLOP barrier

There's a race to deploy the first supercomputer in the U.S. to break the exaFLOP barrier. Intel says it will be first with Aurora in 2021, while AMD and Cray claim they will be first with Frontier. Either way, the Department of Energy will win, because both computers will be deployed at DOE facilities.

An exaFLOP is one quintillion (1018) floating-point operations per second, or 1,000 petaFLOPS. To match what a one exaFLOP computer system can do in just one second, you'd have to perform one calculation every second for 31,688,765,000 years.

While the supercomputing stalwarts continue to build their systems, Folding@Home just crossed the exaFLOP barrier ahead of IBM, Intel, Nvidia, and the Department of Energy.

To read this article in full, please click here


Thanks to Andy Patrizio (see source)

COVID-19 to delay enterprise 5G adoption

Enterprise adoption of 5G will be delayed because the 3GPP standards body has been sidelined by the COVID-19 outbreak and can’t finish its work on a critical standard for at least two months, according to ABI Research.

The standard in question, known as Release 16, addresses latency standards and mandating 99.999% uptime, said ABI Research analyst Leo Gergs. It was due for release in March and will now be delayed until at least June.

“Considering that an hour of machine downtime costs each manufacturer on average $300,000, this underlines the importance a high reliability on availability of the communication network,” he said.

To read this article in full, please click here


Thanks to Jon Gold (see source)

BrandPost: 5 Network Security Remedies for Telework

With the COVID-19 (coronavirus) global pandemic, many employers are recommending additional telework to help keep employees safe and productive. More and more, companies are embracing “remote teams” and allowing their employees the opportunity to work from home or telecommute.

Yet home IT devices are still subject to many of the same threats as on-site business devices. Unsecured off-site routers, modems, and other network devices can cause big headaches for employers, and poorly configured home devices can affect the entire organization. They can still be attacked from any device on the Internet, but they are also vulnerable to unauthorized access from neighbors and passersby.

To read this article in full, please click here


Thanks to CIS (see source)

BrandPost: 4 Risks of Waiting to Migrate to the Cloud

If you’re responsible for the security of your organization's digital environment, staying up-to-date with the latest hardware, environment, and software vulnerability patches can be a challenge. Migrating your workloads to the cloud can help address these challenges in new, unique ways. Waiting to migrate to the cloud can create unforeseen consequences. Here are four risks of waiting to migrate to the cloud and how CIS resources can help mitigate them. 

  1. Lack of independent security configurations

By leveraging virtual machines (VMs) in the cloud, systems admins and CISOs can deploy a single image across multiple workstations. Starting with a base image is an option, but base images lack vendor-agnostic security configurations.

To read this article in full, please click here


Thanks to CIS (see source)

BrandPost: 5 Network Security Remedies for Telework

With the COVID-19 (coronavirus) global pandemic, many employers are recommending additional telework to help keep employees safe and productive. More and more, companies are embracing “remote teams” and allowing their employees the opportunity to work from home or telecommute.

Yet home IT devices are still subject to many of the same threats as on-site business devices. Unsecured off-site routers, modems, and other network devices can cause big headaches for employers, and poorly configured home devices can affect the entire organization. They can still be attacked from any device on the Internet, but they are also vulnerable to unauthorized access from neighbors and passersby.

To read this article in full, please click here


Thanks to CIS (see source)