Blog Posts in November 2014
Can You AS-Prepend a Single Host Route?
Someone recently sent me this question:
Is it possible to prepend one IP address from a public IPv4 segment?
I don’t want to know what crazy stunt this engineer was forced to pull off, but just in case you land in a similar quandary here’s how you shoelace yourself out of it.
Thanks for being there!
A third of my readers are celebrating Thanksgiving today, and I’d like to use the opportunity to say what I always wanted to say but somehow never got to it. Let’s make it short: Thank you! Without you, there would be no ipSpace.net.
Transactional Thoughts on a Stormy Night
It was a dark stormy autumn night and three networking engineers had nothing better to do than ponder the heavy topics of transactional consistency in a distributed SDN environment in Episode 16 of Software Gone Wild podcast.
Here are a few of the topics that crossed our minds:
FECs, LDP, and BGP in the MPLS World
After discussing the basics of MPLS and LDP in our chat, Seamus Gilchrist and myself focused on a concept that perplexes many networking engineers entering the MPLS world: the relationship between Forward Equivalence Classes (FEC), LDP and BGP.
heQuick Peek: Juniper vMX Router
While the industry press deliberates the disaggregation of Arista and Cisco, and Juniper’s new CEO, Juniper launched a virtual version of its vMX router, which is supposed to have up to 160 Gbps of throughput (as compared to 10 Gbps offered by Vyatta 5600 and Cisco CSR). Can Juniper really deliver on that promise?
Moving Workloads to the Clouds
David Spark published 16 tips for moving your workloads to the clouds. Contrary to the usual useless nonsense coming down from hybrid cloud evangelists (you know, the people who moved from “VMs following the sun” to “seamless hybrid cloud workload mobility”) some of the tips actually make sense, starting with “Have a real reason for the migration”. Enjoy!
Open vSwitch Performance Revisited
A while ago I wrote about performance bottlenecks of Open vSwitch. In the meantime, the OVS team drastically improved OVS performance resulting in something that Andy Hill called Ludicrous Speed at the latest OpenStack summit (slide deck, video).
Let’s look at how impressive the performance improvements are.
Network Programmability Phase 2: the Provisioned Network
After describing the current state of affairs in his Network Programmability 101 webinar, Matt Oswald moved to the low-hanging fruits: automating repetitive tasks in baby steps, from VLAN provisioning to consistent device configurations.
Published on , commented on March 10, 2023
Coping with Byzantine Routing Failures
One of my readers sent me an interesting challenge:
We have two MPLS providers sending us default routes and it seems like whenever we have problem with SP1 our failover is not happening properly and actually we have to go in manually and influence our traffic to forward via another path.
Welcome to the wondrous world of byzantine routing failures ;)
Do We Have Too Many Knobs?
The last day of Interop New York found me sitting in the Speaker Center with a few friends pondering the hype and reality of SDN and brokenness of traditional network products. One of the remarks during that conversation was very familiar: “we have too many knobs to configure”, and I replied “and how many knobs do you think there are in Windows registry?" (or Linux kernel and configuration files).
Just Published: Overlay Virtual Networks in Software Defined Data Centers
Overlay virtual networks are one of my favorite topics – it seems I wrote over a hundred blog posts describing various aspects of this emerging (or is it reinvented) technology since Cisco launched VXLAN in 2011.
During the summer of 2014 I organized my blog posts on overlay networks and SDDC into a digital book. I want to make this information as useful and as widely distributed as possible – for a limited time you can download the PDF free of charge.
Reinventing the wheel (or RFC 1925 sect 2.11)
Simon Wardley is another old-timer with low tolerance for people reinventing the broken wheels. I couldn’t resist sharing part of his blog post because it applies equally well to what we’re seeing in the SDN world:
No, I haven't read Gartner's recent research on this subject (I'm not a subscriber) and it seems weird to be reading "research" about stuff you've done in practice a decade ago (sounds familiar). Maybe they've found some magic juice? Experience however dictates that it'll be snake oil […]. I feel like the old car mechanic listening to the kid saying that his magic pill turns water into gas. I'm sure it doesn't ... maybe this time it will ... duh, suckered again.
Meanwhile the academics already talk about SDN 2.0.
Viptela SEN: Hybrid WAN Connectivity with an SDN Twist
Like many of us Khalid Raza wasted countless hours sitting in meetings discussing hybrid WAN connectivity designs using a random combination of DMVPN, IPsec, PfR, and one or more routing protocols… and decided to try to create a better solution to the problem.
Just Published: VXLAN 2.0 Videos
Last week I ran the second part of the updated (4-hour) VXLAN webinar. The raw videos are already online and cover these topics:
- VXLAN-related technologies, including encapsulation, IP multicast use, unicast VXLAN, and VXLAN-over-EVPN;
- VXLAN implementations, including Cisco Nexus 1000v, VMware vCNS, VMware NSX, Nuage VSP and Juniper Contrail;
- VXLAN gateways, including Arista, Brocade, Cisco and Juniper;
- Hardware VTEP integration with OVSDB and EVPN;
- VXLAN-based data center fabrics, including Cisco’s ACI.
Scaling the Cloud Security Groups
Most overlay virtual networking and cloud orchestration products support security groups – more-or-less-statefulish ACLs inserted between VM NIC and virtual switch.
The lure of security groups is obvious: if you’re willing to change your network security paradigm, you can stop thinking in subnets and focus on specifying who can exchange what traffic (usually specified as TCP/UDP port#) with whom.
Handling the Bottom of MPLS Stack
MPLS bottom-of-stack bit confused one of my readers. In particular, he had a problem with the part where the egress MPLS Label Switch Router (LSR) should go from labeled (MPLS) to unlabeled (IPv4, IPv6) packets and had to figure out what was in the packet.
Response: Midokura Open-Sources MidoNet
Last week Midokura decided to follow Juniper’s lead and make MidoNet an open-source product (after all, a similar move worked really well for Contrail, right?).
I may be too skeptical (again), but I fail to see how this could possibly work.
FlipIT Cloud: Orchestrating IT-as-a-Service on Software Gone Wild
Imagine being an IT administrator running a multi-tenant enterprise environment (example: an SMB business center). How many things would you have to configure to add a new tenant? How about adding a new user for an existing tenant?
The engineers behind the scenes of FlipIT cloud service ended up with a 40-page configuration guide when they started the service years ago… and obviously decided full-blown automation is the only way to go.
So You’re an Open Source Shop? Really?
I carried out an interesting quiz during one of my Interop workshop:
- How many use Linux-based servers? Almost everyone raised their hands;
- How many use Apache or Tomcat web servers? Yet again, almost everyone.
- How many run applications written in PHP, Python, Ruby…? Same crowd (probably even a bit more).
- How many use Nginx, Squid or HAProxy for load balancing? Very few.
Is there a rational explanation for this seemingly nonsensical result?
Does a Cloud Orchestration System Need an Underlying SDN Controller?
A while ago I had an interesting discussion with a fellow SDN explorer, in which I came to a conclusion that it makes no sense to insert an overlay virtual networking SDN controller between cloud orchestration system and virtual switches. As always, I missed an important piece of the puzzle: federation of cloud instances.
2014-11-04 16:48Z: CJ Williams sent me an email with information on SDN controller in upcoming Windows Server release. Thank you!
Use a Disaster Recovery Project to Build Your New Cloud
It doesn’t make sense to build a new data center network to support legacy bare-metal server infrastructure. You’ll have to use relatively expensive 1G/10G ports to be able to connect the current and future servers, and once the server and virtualization engineers wake up and do hardware refresh you’ll end up with way too many ports (oh, and you do know that transceivers could cost more than the switching hardware, right?).