In 2014, we did a series of podcasts on Snabb Switch (Snabb Switch and OpenStack, Deep Dive), a software-only switch delivering 10-20 Gbps of forwarded bandwidth per x86 core. In the meantime, Snabb community slowly expanded, optimized the switching code, built a number of solutions on top of the packet forwarding core, and even forked a just-in-time Lua compiler to get better performance.
To find out the details, listen to Episode 91 of Software Gone Wild in which Luke Gorrie explained how far the Snabb project has progressed in the last four years.
BDS: A centralized near-optimal overlay network for inter-datacenter data replication Zhang et al., EuroSys’18
(If you don’t have ACM Digital Library access, the paper can be accessed either by following the link above directly from The Morning Paper blog site).
This is the story of how inter-datacenter multicast transfers at Baidu were sped-up by a factor of 3-5x. That’s a big deal!
For large-scale online service providers, such as Google, Facebook, and Baidu, an important data communication pattern is inter-DC multicast of bulk data — replicating massive amounts of data (e.g., user logs, web search indexes, photo sharing, blog posts) from one DC to multiple DCs in geo-distributed locations.
To set the scene, the authors study inter-DC traffic at Baidu over a period of seven days. Nearly all inter-DC traffic is multicast (91.1%), highlighting the importance of optimising the multicast use case.
When looking at the individual transfers, there is great diversity in the source and destination DCs. Thus it’s not going to suffice to pre-configure a few select routes: “we need a system to automatically route and schedule any given inter-DC multicast transfers.”
60% of the transferred files are over 1TB Continue reading
IHS Markit also reported drops in SD-WAN revenue for Huawei and InfoVista, but these could simply be reflections of quarterly variations in contracts.
If you are considering an SD-WAN or if you already have implemented the technology, let us know what challenges you are facing and what innovations are most compelling.
AT&T said it will still use its own cloud-based connectivity platform but it will work with WING so IoT customers have a single portal to manage their devices.
Newly discovered capabilities mean hackers can snoop on web traffic and execute man-in-the-middle attacks before killing the device.
Having digital twins that learn at the edge can help ease concerns over data custody, confidentiality, and privacy as major cloud vendors deliver more machine learning-based insights to the edge.
The fine is set to resolve ZTE's trade violations, but could open the door for China's approval of Qualcomm's pending NXP acquisition.
The managed SD-WAN service allows BT’s enterprise customers to design, deliver, and evolve their SD-WAN. It improves upon BT's previous attempt to bootstrap old Cisco routers and IWAN to create an SD-WAN overlay.
A quick and dirty post on running the Juniper vQFX on VMWare ESXi.
You might be wondering why ESXi seeing as we’re all cloudy types. ESXi is purely a case of laziness. Each server in my control has ESXi 6.5 installed. This becomes tin management at the most basic level.
Part of my home network has a DMZ which has several public IP addresses and I expose systems and VNFs externally over the internet. More recently thanks to the IP fabric craze, part of what I’m exploring is easy integration and feature enhancement on Juniper vQFX instances. Two choices exist:
I went for the easy ground (because why make it harder than it has to be?) Turns out, it wasn’t as straight forward as it should be, although not difficult. Just a niggle.
Grab yourself the RE and PFE images from the Juniper download site:
https://www.juniper.net/support/downloads/?p=vqfxeval I Grabbed the 18.1 RE and the 17.4 PFE image.
Next, extract the two
.vmdkfiles from the
.boxfiles. You can use the trusty tar tool to extract the files required. Below are two files Continue reading
Enterprise customers often ask the Ansible Network team about the most common use cases for network automation. For this blog post I want to talk about one of the most used (and most versatile) set of network modules: the command
modules. The command modules let you run networking commands with Ansible, the same way a network engineer would type them on the command line. With Ansible, though, the output doesn’t just fly by the terminal window to be lost forever; it can be stored and used in subsequent tasks. It can also be captured in variables, parsed for use by other tasks, and stored in host variables for future reference.
Today we’re going to cover basic use of the network command
modules, including retaining command output with the register
parameter. We’ll also cover scaling to multiple network devices with hostvars
and adding conditional requirements with the wait_for
parameter and three related parameters: interval
, retries
, and match
. The takeaway from this blog post is that any repeatable network operations task can be automated. Ansible is more than configuration management, it allows network operators the freedom to decouple themselves from routine tasks and save themselves time.
There are command modules Continue reading