One of my readers watched my Leaf-and-Spine Fabric Architectures webinar and had a follow-up question:
You mentioned 3-tier architecture was dictated primarily by port count and throughput limits. I can understand that port density was a problem, but can you elaborate why the throughput is also a limitation? Do you mean that core switch like 6500 also not suitable to build a 2-tier network in term of throughput?
As always, the short answer is it depends, in this case on your access port count and bandwidth requirements.
Read more ...This blog has been quiet since my last post in November 2016, covering the announcement that Broadcom is acquiring Brocade, and selling off my part of the business. That was over four months ago, and many of you will be wondering what’s happening. Unfortunately I have no real news: we still don’t know what’s happening.
Originally we were unsure if the IP business would be sold in whole, or broken into parts. We can now see that it is being broken into parts: Arris is acquiring the Ruckus Wireless and ICX Switch business unit. That does not include my part of the business.
Broadcom is continuing to seek buyers for my business unit (Data Center IP, covering SLX, MLX and VDX product families, and of course StackStorm). They are also looking for buyers for the Software Business Unit (vRouter, SDN Controller, and vADC). There are no published timetables for when this process will be complete: it will be done when it’s done.
This means that I still don’t know what’s going to happen to me. My visa is tied to my employer. A change in employer could mean I have to leave the United States. Continue reading
This blog has been quiet since my last post in November 2016, covering the announcement that Broadcom is acquiring Brocade, and selling off my part of the business. That was over four months ago, and many of you will be wondering what’s happening. Unfortunately I have no real news: we still don’t know what’s happening.
Update: We now have news: It’s Extreme Networks
Originally we were unsure if the IP business would be sold in whole, or broken into parts. We can now see that it is being broken into parts: Arris is acquiring the Ruckus Wireless and ICX Switch business unit. That does not include my part of the business.
Broadcom is continuing to seek buyers for my business unit (Data Center IP, covering SLX, MLX and VDX product families, and of course StackStorm). They are also looking for buyers for the Software Business Unit (vRouter, SDN Controller, and vADC). There are no published timetables for when this process will be complete: it will be done when it’s done.
This means that I still don’t know what’s going to happen to me. My visa is tied to my employer. A change in Continue reading
If you’re familiar with Linux, you know how important and exciting it can be to submit new technology that is accepted into the kernel. If you’re not familiar with Linux, you can take my word for it (and I highly suggest you attend one of our bootcamps). Many networking features are motivated by an OS for switches and routers, but most if not all of those features prove useful for other use cases as well. Cumulus Networks strives for a uniform operating model across switches and servers, so it makes sense for us to spend the time and effort getting these features into upstream code bases. An example of this effort is VRF for Linux.
I joined Cumulus Networks in June 2015 to work on a VRF solution for Linux —to create an implementation that met the goals we wanted for Cumulus Linux and was acceptable to upstream maintainers for Linux as a whole. That solution was first available last year with Cumulus Linux 3.0 and because of the upstream push that solution is rolling out in general OS distributions such as Debian Stretch and Ubuntu 16.
This post is a bit long, so I start with a high Continue reading