Tuesday, November 8, 2022

Uselessness of the Network Diagrams

Network Engineers reading Network Diagrams which are mostly:

Dated
Half baked
Capturing only what the document writers feels about " should be good enough "
Are useless in 90% cases
Have no orientation towards the intended audience " Whom it is meant for, What situations those will be used in, What the workflow would look like, What is the most important set of information we need to capture"
Have no " Quality Assurance " process in place
Serves as information fillers to maintain the status quo and useless process

HTH...

A Network Artist 🎨

Monday, October 31, 2022

Marrying SASE (SDWAN) with 5G - The Marketing, The Myths & The Fallacies & How to Get it Right (An Architectural Perspective)

When almost 3 years ago I wrote about why having an inbuilt LTE interface inside a SD-WAN device doesn't really matter, the 5G thingy was still relatively new.

During a recent Enterprise Architecture Consulting engagement, I was asked by one of my client if they should really care about 5G and 5G interfaces on the variety of SD-WAN platforms that were pitched to them by different Systems Integrators (SIs) & MSP (Managed Service Providers)/Telcos.

So let's start with a simple question - "What problem we are trying to solve?"

In general you will see a few types of customers in SASE/SD-WAN market :

1. Which have Technical/Solutions architects those are completely sold on vendors marketing (50% of the crowd)

2. Those who wants to jump on the bandwagon due to fear of being left behind in the similar industry or by the competition (25% of the crowd)

3. Those who always are either too excited by technology or have a lot of money to throw onto the problem (the next 20%)

4. Those who can really map business capabilities to technology capabilities (the rare and the last 5%)

In general, you would often find a few ways the 5G gets included into the solution by solution providers such as :

Design 1 - You have a site (mid/large size) which either has got hybrid connectivity (MPLS + Internet) or 2 x Internet connects, while keeping 5G cellular as a last resort backup link in an event of a total failure.

Design 2 - A small site that usually runs on a single internet link and keeping 5G as back for last resort.

Design 3 - 5G as backup of last resort onto your DC WAN edge

Now in general it doesn't look like a bad idea to have a 5G interface but here are few of the important considerations:

- In general your DCs/COLOs and Campus networks would be setting on racks behind thick physical building structures (remember your Wi-Fi coverage problems even while your APs are sitting inside), so very often you would expect coverage issues. Now one of the argument here might be that I can install an external antenna of some sort on top of building or floor and run a fiber cable connection from there. Fair...but:

> Now you got to get approval for installations, cable runs, have administrative processes and safety processes in place and what not (What if the lightening hit the antenna?).

- To my understanding, most SASE/SD-WAN solutions don't offer any visual monitoring, reporting & troubleshooting tools for - Checking 5G signal strength, 5G interface troubleshooting, Dummy traffic probes etc.

- Interestingly enough now you need even a more complex traffic distribution, traffic prioritization, traffic failover, traffic desired SLA/performance metrics and other set of policies into the mix. And even if you end up doing that successfully, how you are going to document it for the operations ?, Is your EMS/NMS equipped with such capabilities ?

- From the network architecture perspective, you just added an another layer of complexity. Assuming this 5G interface is an HW module, you got to now deal with: New stack of software and protocols within your fancy WAN edge device (3GPP standards), New interaction surfaces, New potential grey failures.

- You just end up adding the more state into the network (State, Surface & Optimization tradeoffs)

- You need to now have life cycle mgmt. in place for your 5G interface (HW/SW Upgrades, Monitoring, Management, Refresh etc.) beside that fact that the 5G specifications often vary country to country (even from Telco to Telco) and now you need to keep track of Data Plans, Data Usage, Availability & Performance Mgmt., Cost Mgmt. and what not. (Remember your data plans are pretty limited in general?)...imagine to solve these problems at a global scale deployment dealing with different MSP.

> How to you move to 6G if it comes out in the next few years ?

Now after all these interesting questions, we may still ask:

1. What are the better alternatives today ?

2. Where 5G might still make sense ?

Answering the first question, IMHO I would still recommend you to opt for a broadband connection and avoid 5G unless you have a very particular problem or scenario because:

- With broadband you are still dealing with Ethernet connection between your CPE and broadband router/device which is a pretty familiar connectivity model and protocol stack to deal with.

- In general your broadband data plans are much bigger

- You don't need to deal with another MSP for service management perspective as in general your ISP would have a common portal to give you a view of all MPLS, Enterprise Class Internet and Broadband based Internet connections. (Also mind that historically your SPs were building two parallel networks for ISP (Internet Service Provider) and MSP (Mobile Service Provider) business units, though those are converging now more and more)

- With everyone hit by pandemic that accelerated WFH culture, in both developing and developed countries you would expect to have broadband being available very easily and at the affordable prices. The only places you might still face availability issues are Tier-3 cities and so forth. But for most part that is not a technology problem but your SPs wanting you to stick with cellular connectivity to drive more profits (So its an intent problem rather)

- My own research suggests that in many developed countries getting a broadband will cost you far less compare to the enterprise class 4G/LTE/5G/Private 5G connection

Answering the second question, here are few use cases for your consideration:

- 5G as part of your SD-WAN transition plan (Legacy to SD-WAN)

- Your last mile hybrid or other wired connectivity model still runs on same fiber or shared media (true path diversity problem)

- One or both of you last miles are on Microwave (Should be pretty rare now)

- 5G as a OOB (Out of Band) management option

- Movable workplace and offices (Eg. Marketing/Sales & Promotion offices or moving semi-trailer trucks often used in variety of businesses)

- Specific operating conditions such as in Oil/Gas & Mining industries

- Edge compute/Cloud

- IoT Platforms

HTH...

A Network Artist 🎨

Further Readings:

SD-WAN Leads to $96,000 4G/LTE Bill

Improve Your Home Internet Performance Using CoDel

More Bandwidth Doesn’t Matter (much)

Are Networks Really Complex ?

Enterprise QOS Design & Deployment - Good, Bad Or Ugly ?

Focus on Your Business, Not Fancy Technologies

Are You Solving the Right Problem?

This Is What Makes Networking So Complex

Are Business Needs Just Excuses for Vendor Shenanigans?

The Three Paths of Enterprise IT

SDN Will Not Solve Real-Life Enterprise Problems

Why Intent Based Networking (IBN) will Not Save Your Network Anytime Soon ?

Complexity and the Thin Waist

It’s Most Complicated than You Think

Details and Complexity

Monday, May 16, 2022

How Many Routes My ASIC Can Hold ? - A Short Post

In late 2021 I wrote this brief article around skills a network engineer should pick on in his/her early career. Where I suggested to be at least familiar with basic understanding of both "Router Architecture" & "ASIC Architecture". Now obviously the depth is always subjective to:

1. How many details I need to know for my current role & responsibilities in order to get things right

2. The amount of details and depth I need to know for potential future roles that you might be targeting

3. Your personal curiosity & interest

4. If you are really into Architecture & Design, You got to have fair & intermediate level of understanding of these topics at minimal

5. If you into Pre-sales, You often got to deal with sizing & performance for a given set of equipment as part of your deliverables requested by client in the form of RFP or RFI. Remember those data sheets you often have to refer to claiming IPv4 or IPv6 prefixes numbers a platform can hold/support ?

6. You might have to do platform testing at some point as part time or full time job including you may land yourself into a COE (Centre of Excellence) of your organization or may end up into a "Platform/Service Product Management" role.

Now you must know that often those specifics and details are hidden and never publicly shared/offered by most of ASIC/Platform vendors. You often got to be a premium customer and sign-off tons of NDAs to get those details to some extent and more importantly you got to be very specific around what exactly you are looking for since asking for data in abstraction would often result into tons of non-specific information thrown on you by your fav. ASIC vendor.

Assuming by now you have some more clarity in terms of why you need to know all these details as a Network engineer depending upon where you are and where you plan to end up, lets circle back to original topic for today which is "How many IPv4 (could be IPv6) prefix my device support in reality?"

Which leads us to a simple question - "What are the different variables I am dealing with when trying to figure how much routes my platform can really hold?"

While a simple answer you would often hear would be "it depends" or someone may point your to RFC-1925 rule 8 "It is more complicated than you think"

So let's try to list some of them in this series Part - 1