Research: BBR: Congestion-Based Congestion Control – ACM Queue
The BBR algorithm appears to be building critical mass of support in the Internet community which makes reading this research paper even more worthwhile.
When bottleneck buffers are small, loss- based congestion control misinterprets loss as a signal
of congestion, leading to low throughput. Fixing these problems requires an alternative to loss-based congestion control. Finding this alternative requires an understanding of where and how network congestion originates.
BBR: Congestion-Based Congestion Control – ACM Queue : http://queue.acm.org/detail.cfm?id=3022184
The post Research: BBR: Congestion-Based Congestion Control – ACM Queue appeared first on EtherealMind.
NANOG 69
NANOG 69 was held in Washington DC in early February. Here’s my notes from the meeting.Opinion – Vendors Please Stop the Mud Slinging!
I’m not known for going on rants but lately I’ve been seeing a lot of stupid tweets from vendors that have really bothered me. So today I’ll give my best Tom Hollingsworth “networkingnerd” impression and tell you what’s on my mind. To give you an example what the vendor marketing teams are putting out there I give you this piece of work:
At first it seems a bit cute and funny. Oh look! It’s Star Wars! All nerds love Star Wars! I do too, just to be clear. What this kind of marketing does though is to dumb down the customers. It insults my intelligence as a Network Architect. Hardware still matters. There still is a physical world. Almost all projects in networking has some kind of existing network so almost all deployments are going to be brownfield to some extent. Please show me the organization that does not have an existing network and is going to deploy something like NSX or ACI for their first network. Please show me the organization that has no legacy systems or applications. Please show me the organization that develops and owns all of their applications and they are all nicely Continue reading
Intel Gets Serious About Neuromorphic, Cognitive Computing Future
Like all hardware device makers eager to meet the newest market opportunity, Intel is placing multiple bets on the future of machine learning hardware. The chipmaker has already cast its Xeon Phi and future integrated Nervana Systems chips into the deep learning pool while touting regular Xeons to do the heavy lifting on the inference side.
However, a recent conversation we had with Intel turned up a surprising new addition to the machine learning conversation—an emphasis on neuromorphic devices and what Intel is openly calling “cognitive computing” (a term used primarily—and heavily—for IBM’s Watson-driven AI technologies). This is the first …
Intel Gets Serious About Neuromorphic, Cognitive Computing Future was written by Nicole Hemsoth at The Next Platform.
Kubernetes CRI and Minikube
Kubernetes CRI(Container runtime interface) is introduced in experimental mode in Kubernetes 1.15 release. Kubernetes CRI introduces a common Container runtime layer that allows for Kubernetes orchestrator to work with multiple Container runtimes like Docker, Rkt, Runc, Hypernetes etc. CRI makes it easy to plug in a new Container runtime to Kubernetes. Minikube project simplifies Kubernetes … Continue reading Kubernetes CRI and Minikube
NSX-V 6.3: Cross-VC NSX Security Enhancements
NSX-V 6.2 introduced the Cross-NSX feature to allow for NSX logical networking and security across multiple vCenter domains. The ability to apply consistent networking and security across vCenter domains provides for mulitple use cases for Cross-VC NSX: workload mobility, resource pooling, multi-site security, ease of automation across sites, and disaster avoidance/recovery. With the recent release of NSX-V 6.3, several enhancements have been added to the Cross-VC NSX feature to provide for additional capabilities and overall robustness of the solution. In this blog post I’ll discuss the new Cross-VC NSX security enhancements in NSX-V 6.3. For additional information on Cross-VC NSX check-out my prior Cross-VC NSX blog posts.
The security enhancements for Cross-VC NSX can be grouped into two categories:
- General Enhancements (Apply Across both Active/Active and Active/Standby deployment models)
- Enhancements for Active/Standby Use Case
Active/Active and Active/Standby above refers to if the application is active at both sites or if it is active at one site and standby at another site (ex: disaster recovery). Enhancements for both of these respective categories are discussed in more detail below.
1.) General Enhancements (Apply Across both Active/Active and Active/Standby deployment models)
Figure 1: Cross-VC NSX Active/Standby and Continue reading
Cisco’s Former Spin-In Crew Resurfaces
What's Pensando working on? Not hard to guess
Worth Reading: Dynamics on emerging spaces
The post Worth Reading: Dynamics on emerging spaces appeared first on 'net work.
Using oVirt and Vagrant
Introducing oVirt virtual machine management via Vagrant.
In this short tutorial I'm going to give a brief introduction on how to use vagrant to manage oVirt with the new community developed oVirt v4 Vagrant provider.
Background
Vagrant is a way to tool to create portable and reproducible environments. We can use it to provision and manage virtual machines in oVirt by managing a base box (small enough to fit in github as an artifact) and a Vagrantfile. The Vagrantfile is the piece of configuration that defines everything about the virtual machines: memory, cpu, base image, and any other configuration that is specific to the hosting environment.
Prerequisites
- A fully working and configured oVirt cluster of any size
- A system capable of compiling and running the oVirt ruby SDK gem
- Vagrant 1.8 or later
- The oVirt vagrant plugin installed via
$ vagrant plugin install vagrant-ovirt4
The Vagrantfile
To start off, I'm going to use this Vagrantfile:
Vagrant.configure("2") do |config|
config.vm.box = 'ovirt4'
config.vm.hostname = "test-vm"
config.vm.box_url = 'https://github.com/myoung34/vagrant-ovirt4/blob/master/example_box/dummy.box?raw=true'
config.vm.network :private_network,
:ip => '192.168.56.100', :nictype => 'virtio', :netmask Continue readingAppDynamics Acquisition Sets the Stage for Cisco in the APM Market
But couldn't Cisco have spent less than $3.7B? Questions remain.