As I said, the Cisco Catalyst 9300 is a switch family that I am familiar with due to my current work environment. I have deployed all different models from C9300-48U-A, C9300-48UN-A, to the C9300-48UXM-A. They have been a reliable switch from my experience, albeit a bit of pain due to the interface naming differences between the models. ( that is another blog post altogether )

I am honored to be welcomed back to Tech Field Day Exclusive with Cisco scheduled on 16-17 March 2022. It will be a live-streamed event that you can watch over on the Tech Field Day site during the event, or on their YouTube channel shortly after.

The schedule for the event is as follows:

See you there!

I have one way to describe the Broadcom session at NFD22, that term is NERDGASM! The presenters had to be some of the guys who are designing these chips, the depth of knowledge they had was incredible.

Broadcom are the engineers that make the merchant silicon chips that manufacturers like Cisco, Juniper, Arista, etc all use in their hardware. As you can see from the slide above, Broadcom is the heart of the unit as it is the silicon inside the switch/router.

Recently I attended Networking Field Day 22 (NFD22) in San Jose, CA as a delegate. By attending NFD you may be exposed to companies that you may not normally come across. The company that I am going to talk about here is called Gluware and more specifically about how they can help with keeping your network configurations in compliance with the desired state.

What Gluware is wants to do you for your business is be the solution that can solve the automation and configuration compliance challenges. Their mission is “to apply powerful intelligence to enable transformative automation.” They also did state that they are a multivendor and multidomain ( LAN / WAN / Wireless) solution, which in my experience is important as multi-vendor networks are common.

There was something they mentioned that resonated with me. It was how they summed up the current state of automation into three main parts:

Recently I attended Networking Field Day 22 (NFD22) in San Jose, CA as a delegate. By attending NFD you may be exposed to companies that you may not normally come across. The first company that presented, DriveNets, was one of those companies and here is their story.

DriveNets was founded a little over 4 years ago with a mission to “disrupt network economics with growth-ready cloud-based infrastructure.” Their main target market is the Tier 1 ISP ( ATT, NTT, Level3, etc). They started with four things in mind.

1. Software-Based
2. Scale
3. Open Architecture
4. Simple Operation

When you look at a “legacy service provider” network, you will see that there are four primary layers. You have

• Last Mile (CE Router)
• Metro Aggregation (Metro)
• Provider Edge (PE Router)
• Core (P Router)

QFX 10000

 

The QFX10000 is Juniper Networks latest addition to Data Center networking and it comes with Juniper Networks latest purpose-built ASIC -the Q5. The Q5 chip enables unprecedented performance, port density, and scale as can be seen in the QFX 10000.  What is also very interesting is that the Q5 chip uses Memory Cube technology, something that is not found in merchant silicon.  The QFX10000 is a high density spine switch optimized for the Data Center (or as many call it, “the cloud.”)  How high density?  If you look at the image above you will see an image of the QFX100002–72Q switch.  Those are 72 –  40 GbE ports in that 2U switch!  That equates to 288 –  10 GbE ports! Not enough bandwidth for you?  Well, you can also order these boxes with  24 –  100 GbE QSFP28 interfaces instead!  That is some serious bandwidth options right there!  Still not enough for you? You can get these switches in a chassis design –  8–slot and 16–slot chassis options are available.  The 8–slot chassis will support up to 1152 –  10GbE ports and the 16–slot will support up to 2304 –  10GbE ports!  You can find out more about the QFX 10000 here –  JUNIPER QFX10000.  For more information on the Memory Cube / 3-D Memory technology, there is an great post by Chang-Hong Wu on the Juniper forums – 3-D Memory.

November was a busy month for me on the job front as well as personal vacation.  What also happened is that I was a paid Solar Winds Thwack Ambassador for the month (Nov 2013) as well and posted some fun and intersting items.
 
I wanted to take a moment and share them with you in case you did not get a chance to view them prior
 
My first post was Network Topology Horror Stories –  http://thwack.solarwinds.com/thread/60335
There are some great stories in this post, just wow sometimes!
 
The second post was Application Intelligence in the Network –  http://thwack.solarwinds.com/thread/60485
This post was right after the Cisco, Arista, and Juniper announcements in November.
 
Third was What Tips and Tricks do you have for Networking –  http://thwack.solarwinds.com/thread/60838
Some great ideas in this post, really worth checking out!
 
And my final post for the month was What Network Support and Troubleshooting Are in your Software Toolkit? –  http://thwack.solarwinds.com/thread/60636
Again, some great ideas and software tools to check out!
 
Check them out and share your information and stories if you like!
 
 
 
 

On October 26th I attended the OpenFlow Symposium in San Jose, CA hosted by Packet Pushers and TechFieldDay at the DoubleTree San Jose. The purpose of this symposium was to help gain more exposure to OpenFlow as well as discuss what some of the major players see and feel about the technology. In attendance was representatives from Big Switch Networks, Brocade, Cisco, Google, Juniper, NEC, and Yahoo!. They each were allowed to present their views and feelings and then the panel was opened up for discussion and questions.
My take was that Google and Yahoo! are big into this technology as they have the most to gain from it.  They run warehouse size data centers all over the world, and this type of technology has a huge benefit to them as it means they can develop their own controllers and control merchant silicon, thus releasing them from relying on other vendors network hardware to support their data center. Not only that, but they can help prioritize and track individual flows based on what the user is doing. For example – a person who is placing items in their carts and heading to checkout can have a higher priority flow over someone just looking around.  The speed to finish the transaction is important to them, not to mention they can also secure that traffic in the network differently then a person who is just looking.

An OpenFlow Symposium was held in San Jose, CA on October 26th, 2011 and hosted by by PacketPushers.Net and TechFieldDay.Com. This was my first true exposure to the OpenFlow protocol and concepts of SDN (Software Designed Networking), and it was a great exposure.
If you are not familiar with OpenFlow or SDN, please let me take a moment and give you a brief overview and a quick scenario. OpenFlow is a way in which you can have a application – called a Controller – maintain and control switch flow tables. This controller is aware of all flows in the topology of its domain, the domain being the switches/ports that it controls. Yes, you can have specific ports on a switch controlled via the controller, and all the other ports as part of a normal network that is controlled via the switch. OpenFlow will have no impact to the other ports, they will continue to operate just like nothing was different.
A good use case for OpenFlow is Multi-Path Layer 2 switching in a data center or other high bandwidth location. With normal multi-patch technologies one of the links is blocked via STP, thus sitting idle just waiting for the primary link to fail. With OpenFlow, the controller is aware of all paths in the network and able to load the traffic accordingly. Lets take a look at the following diagram: