Stumbled upon this via HighScalability:
Every time I feel like I'm "out of touch" with the hip new thing, I take a weekend to look into it. I tend to discover that the core principles are the same [...]; or you can tell they didn't learn from the previous solution and this new one misses the mark, but it'll be three years before anyone notices (because those with experience probably aren't touching it yet, and those without experience will discover the shortcomings in time.)
Yep, that explains the whole centralized control plane ruckus ;) Read also a similar musing by Ethan Banks.
We did several podcasts describing how one could get stellar packet forwarding performance on x86 servers reimplementing the whole forwarding stack outside of kernel (Snabb Switch) or bypassing the Linux kernel and moving the packet processing into userspace (PF_Ring).
Now let’s see if it’s possible to improve the Linux kernel forwarding performance. Thomas Graf, one of the authors of Cilium claims it can be done and explained the intricate details in Episode 64 of Software Gone Wild.
Read more ...After finishing the network automation part of a recent SDN workshop I told the attendees “Vote with your wallet. If your current vendor doesn’t support the network automation functionality you need, move on.”
Not surprisingly, the next question was “And what shall we ask for?” Here’s a short list of ideas, please add yours in comments.
Read more ...One of my readers sent me this question:
I often see designs involving several more than 2 DCs spread over different locations. I was actually wondering if that makes sense to bring high availability inside the DC while there's redundancy in place between the DCs. For example, is there a good reason to put a cluster of firewalls in a DC, when it is possible to quickly fail over to another available DC, as a redundant cluster increases costs, licenses and complexity.
Rule#1 of good engineering: Know Your Problem ;) In this particular case:
Read more ...The featured webinar in October 2016 is the Designing Active-Active and Disaster Recovery Data Centers webinar, and the featured videos include the discussion of disaster avoidance challenges and the caveats you might encounter with long-distance vMotion. All ipSpace.net subscribers can view these videos, if you’re not one of them yet start with the trial subscription.
As a trial subscriber you can also use this month's featured webinar discount to purchase the webinar.
One of my readers sent me an interesting question after reading my ICMP Redirects blog post:
In Cisco IOS, when a packet is marked by IOS for ICMP redirect to a better gateway, that packet is being punted to the CPU, right?
It depends on the platform, but it’s going to hurt no matter what.
Read more ...We got pretty far in our Data Center optimization journey. We virtualized the workload, got rid of legacy technologies, and reduced the number of server uplinks and replaced storage arrays with distributed file system.
Final step on the journey: replace physical firewalls and load balancers with virtual appliances.
My friend Matthias Luft sent me an interesting tweet a while ago:
@ioshints What’s your take on firewall rule sets & IP addresses vs. hostnames?
— Matthias Luft (@uchi_mata) August 16, 2016
All I could say in 160 characters was “it depends”. Here’s a longer answer.
Read more ...Just wanted to point you to two excellent blog posts recently published by Russ White.
Reaction: DevOps and Dumpster Fires
If teaching coders isn’t going to solve the problem, then what do we do? We need to go to where the money is. Applications aren’t bought by coders, just like networks aren’t.Read more ...
One of the attendees of my Building Next-Generation Data Center course asked this interesting question after listening to my description of differences between Chet/Puppet and Ansible:
For Zero-Touch Provisioning to work, an agent gets installed on the box as a boot up process that would contact the master indicating the box is up and install necessary configuration. How does this work with agent-less approach such as Ansible?
Here’s the first glitch: many network devices don’t ship with Puppet or Chef agent; you have to install it during the provisioning process.
Read more ...One of my readers sent me interesting feedback after reading my explanation of why I’d try not to use OSPF as a routing protocol between hosts and ToR switches. He said:
Unfortunately we can’t use BGP because IBM mainframes support only OSPF or RIP, so we decided to use VRFs instead.
Here’s what they did:
Read more ...Marco Canini from UC Louvain is working on an IXP research project focused on bringing privacy guarantees into Internet routing context. They’re trying to understand the privacy considerations of network operators and have created a short survey to gather the initial data.
Researchers from UC Louvain have been involved in tons of really useful projects including BGP PIC, LFA, MP-TCP, Fibbing, Software-defined IXP and flow-based load balancing, so if you’re connected to an IXP, please take your time and fill in the survey.
In March 2016 my friend Matt Oswalt announced a distributed network testing framework that he used for validation in his network automation / continuous integration projects. Initial tests included ping and DNS probes, and he added HTTP testing in May 2016.
The project continues to grow (and already got its own Github and documentation page) and Matt was kind enough to share the news and future plans in Episode 63 of Software Gone Wild.
To ask questions about the project, join the Todd channel on networktocode Slack team (self-registration at slack.networktocode.com)
One of my friends plans to replace existing FabricPath data center infrastructure, and asked whether it would make sense to stay with FabricPath (using the new Nexus 5600 switches) or migrate to ACI.
I proposed a third option: go with simple VXLAN encapsulation on Nexus 9000 switches. Here’s why:
Read more ...One of my readers watched my TCP, HTTP and SPDY webinar and disagreed with my assertion that shaping sometimes works better than policing.
TL&DR summary: policing = dropping excess packets, shaping = delaying excess packets.
Here’s the picture he sent me (watch the video to get the context and read this article to get the background details):
Read more ...One of my readers had problems getting the NFV big picture (and how it relates to SDN):
I find the topic area of SDN and NFV a bit overwhelming in terms of information, particularly the NFV bit.
NFV is a really simple concept (network services packaged in VM format), what makes it complex is all the infrastructure you need around it.
Read more ...One of my friends sent me this question:
How many VM moves do you see in a medium and how many in a large data center environment per second and per minute? What would be a reasonable maximum?
Obviously the answer to the first part is it depends (please share your experience in the comments), so we’ll focus on the second one. It’s time for another Fermi estimate.
Read more ...Dinesh Dutt started his excellent Docker Networking webinar with introduction to the concepts of microservices and Linux containers. You won’t find any deep dives in this part of the webinar, but all you need to do to get the details you’re looking for is to fill in the registration form.
While we were preparing for the Cumulus Networks’ Routing on Hosts webinar Dinesh Dutt sent me a message along these lines:
You categorically reject the use of OSPF, but we have a couple of customers using it quite happily. I'm sure you have good reasons and the reasons you list [in the presentation] are ones I agree with. OTOH, why not use totally stubby areas with the hosts being in such an area?
How about:
Read more ...One of my long-time regular readers sent me this question:
I was wondering if you have had any interest in putting together an SD-WAN overview/update similar to what you do with data center fabrics where you cover the different product offerings, differentiators, solution scorecard…
That would be a good idea. Unfortunately the SD-WAN vendors aren’t exactly helping.
Read more ...