CloudNets Season 4

Listen on Apple Podcasts
Listen on Spotify

Hi, and welcome back to CloudNets, where networks meet cloud.

And today we’re going to talk about DNOR, the DriveNets Network Orchestrator.

And we have the DNOR expert, Calin.

Hi.

Hi.

Good to see you, man.

Pleasure to be here.

Okay, so Calin, there are a lot of fun and interesting stuff coming to DNOR in 2024 that will make our audience life easier.

Let’s talk about the three major innovations we have there.

Fantastic.

So DNOR ’24, packed with new automation feature sets.

First up, Autoboot profiles

This helps you bootstrap new white boxes from the factory.

The automation makes it very easy to add white boxes to your network and turn them into fully functioning routers.

And this gives you true zero touch provisioning for your devices.

So a much faster and easier provisioning process altogether.

Absolutely.

Quicker to roll out boxes, faster to deploy a network.

Good.

What else?

Next up is Smart Rollout.

We think this is very innovative.

It’s our brand new feature and it allows you to push software into the entire network after configuring just a few parameters that you’d like it to listen to.

You configure a few parameters and you let DNOR do the rest.

So you configure what you want to do and when you want to do it.

You probably, he knows when are the maintenance windows and he does everything according to plan and you don’t have to worry about it.

Exactly.

It even knows about your maintenance windows.

Cool.

Okay.

And the third thing, because we always have 3 things.

It’s the Single Source of Truth integration that we’re doing.

We’re moving towards Active Inventory.

So we seamlessly integrate the inventory that we have in DNOR into NetBox or other inventory systems.

We push that information up and allows for true automation workflows to take place in between the inventory, what you want the network to look like, and what it looks like today.

Wow.

Okay, so this is cool and this is exciting.

So three things you need and should expect from DNOR in 2024 is:

One Zero touch provisioning or auto boot. That means that your provisioning process is much easier and faster and error-free

The second is Smart Rollout. So basically you tell the system what do you want to do in network and where you want to do it. When are the maintenance windows and it’s scheduled and does the rest.

And the third is Single Source of Truth. That means that you have an active inventory and online truth inventory. And the network is the master instead of inventory being the master and don’t have to manage your network offline. Everything is online and you can be sure that your automation process is hitting the right spots.

So I have to say this is exciting.

So Calin, thank you for bringing this feature and thank you for joining us in CloudNets.

Pleasure to be here and thank you for joining.

See you next time on CloudNets.

Bye.

Listen on Apple Podcasts
Listen on Spotify

Hi, and welcome back to CloudNets,

where networks meet cloud.

And today we’re going to talk about

the road to autonomous network.

And we have our autonomous network and

routing expert, Calin.

Hi, thank you for joining.

Nice to be here.

So Calin, when we envision autonomous networks in the far, far future, we talk about self healing and self configuring networks and some AI entity that will manage it all.

So Calin, is it coming soon?

We’re not there yet as an industry, but it’s in the future and we’re just getting started in this place.

Okay, so no one will give the keys to some AI entities and we’ll all sit home, but we can do something about it right now.

Right, so what are our ideas or vision from DriveNets towards autonomous networks?

Yeah, great.

So what we’re doing is we’re taking baby steps.

We’re going to start with islands of automation and we’re going to automate small, mundane tasks.

Yeah, the boring stuff.

The boring stuff.

And we’re going to do that by helping our operators develop CI/CD pipelines that allow them to take innovation from their lab to a network digital twin and eventually into production in a much faster way.

Okay, so CI/CD basically saves you the need to have a full blown lab and to test any innovation or new feature you have on it, you simply create it in the virtual world like the cloud operators do, and then push it down the network after you’re sure that it works properly.

Exactly.

Network Cloud was born to be in the cloud.

Exactly.
Okay, so this is great.
What else do we work on?
Please let it be related to AI because I need something with AI for the views.
So we’re working on AI Ops.
Oh, to reduce mean time to repair for operators.

How do we do that?

How do we plan to do it?

We’re building a brand new engine.

DriveNets is training our own model that handles and creates root cause analysis with real summarized events for our operators.

So that they basically train the modelon a very large list of alarms and allow it to learn what are the relations between them and how to figure out which alarm is important, what is the root cause analysis exactly.

We take thousands of alarms and we reduce it down to what the operations team needs to focus on in order to resolve the problem. But as we all know, most of the time it takes to repair a problem is to identify what is causing it. Then the rest is usually simple.

Exactly.

Our goal is to become an AI assistant to the operations teams. They can use this as one of their tools in the toolbox when operating their networks.

Great.

And I guess in the future we’ll be able to even reduce, to even increase MTBF because we can know in advance when a fault is going to occur and proactively avoid it.

Exactly.

We’re working on proactive use cases as well to know when something may happen in the network in the future and take steps to prevent it.

Okay, this is cool.

So when you think about autonomous networks, the three things you need to remember is one, it’s not coming soon.

No one will leave the keys to an AI entity anytime soon.

So you need to be prepared, but you don’t need to rush into it.

So the second one is baby step.

The baby steps are islands of automation in which you can implement tasks that will replace the boring stuff you do day after day.

An example is a digital twin that will allow you to test and to develop new features and test it before you push it into the network.

And the third is AIOps. And that means that, for instance, you can do automation on root cause analysis. The AI training procedures learns what is the relation between different alarms and the faults. And you can very fast and you can very quickly identify the alarm and resolve it. In the future, you might be able to foresee the alarm and proactively avoid it.

So this is what’s coming. It’s very exciting and I can’t wait for the future.

Thank you Calin, for telling us about the future.

Thanks for having me here to talk about DNOR and AIOps, and thank you for watching.

See you next time on CloudNets.

Listen on Apple Podcasts
Listen on Spotify

Hi and welcome back to CloudNets, where networks meet cloud.
Today we’re going to talk about DCI, Data Center Interconnect.
No, no, don’t go.
I know it seems like a boring subject, but DCI is going through something and this something is called AI.
And we have Shai, our interconnect and AI expert.
Again, thank you Shai for coming.
Thank. Thank you for having me.

3 changes that AI made to DCI
So Shai, what are three points or three changes that AI made to DCI and what do we need to do with the new DCI requirements?
So we have three things that we need to remember as you said.
First of all, we need to talk about capacity.
Okay.
Secondly, we need to talk about performance.
And thirdly, we need to talk about one environment and layer 3 capabilities.
Okay.
Okay.

1 Capacity
So let’s start with the first one with capacity.
Okay.
We all feel the AI boom that we have right now.
This means that those large scale workloads with thousands of GPUs generate a lot of traffic.
And we need to have DCI solution with enough scalability to handle those traffic flows.
Okay.
This is one no longer single chassis is enough for all CI needs.
You need much more than that.
Okay, what about performance?
Yeah.

2 Performance
Secondly, we have performance.
No gigas.
Performance, AI, great performance.
This means you need a lossless connection between one data center to the other.
So you need deep buffering capabilities in the DCI.
Something that many of the DCI solutions that we have right now doesn’t have.
Okay, we talked about it a bit when we talk about AI workloads and the job completion time performance.
Now it is expanding to the DCI and we feel the heat here as well.

One environment and layer 3 capabilities
Third thing was one environment.
One like layer 3.
You need a router.
Basically.
Yeah.
You need the router capabilities like for example, 1000 EVGPs.
You need to handle those thousands of EVGP connections and you need a real router to handle this.
This means that you need a solution that can handle all those three points capacity.
You need to have something that can handle the lossless connection and have one capability.
Let’s think about solutions.
No such a solution.
I don’t know.
DDC!
No.
Really?
Okay.
So we’ve been talking about it for 4 seasons.
Yes.
But now DDC is a good fit for this.
Yeah.
Imagine that.
Yeah.
Okay, so velocity.
Yeah, it can do it.
It’s in scale, basically.
Yeah.
Secondly, we have the performance, it’s not.
The scheduled fabric rebound offerings.
And what better author do we have than DDC?
The best.
Okay.
Okay.
Wow.
This was an amazing revelation.
So, three things you need to remember about the new DCI.
The DCI in the AI era, the DCI that connects data centers that are moving towards AI.
One is an explosion in capacity, needs no more.
One large, very large chassis, it is not enough anymore.
I think there are some operators that do eight chassis,
and still it is not enough.
So you need limitless capacity.
You need a flawless, lossless performance.
AI performance.
We talked about it a lot.
And you need layer 3 because you need EVGP, et cetera, et cetera.
And all of these exist in the DDC solution available from DriveNets.
So.
Okay, this looks nice.
Okay, thank you very much, and, for joining again.
Thank you for watching.
See you next time on CloudNets.
Bye.

Listen on Apple Podcasts
Listen on Spotify
Watch on YouTube

Hi and welcome to CloudNets where networks meet cloud.
Today I am going to talk about the new service provider network, the network modernization process all service providers need to go to.
Because frankly, we cannot build a network the same way for over 20 years and still maintain a valid business.
So let’s talk about the 3 things you need to know when you build or reinvent your service provider IP and MPLS infrastructure.
Disaggregation
First is disaggregation.
Disaggregation of hardware and software of different network functions of multi level allow you to harmonize the hardware, to use the same type of hardware across your entire network and all network domain and make your network software centric.
This means that you avoid vendor lock.
This means that you can choose any optics, any asic, any ODM that you want and you no longer are dependent on a single vendor.
It also means that you are building your networks like a cloud.
You have cloud like operation, you have simplified significantly the whole operation overhead you have in your network.
And the last is that your innovation, your introduction of new services, is now much faster because it depends on software rather than on software and on hardware.
So first was disaggregations.
Convergence
Second, we’re talking about convergence and I’m talking all types of convergence in your infrastructure, in your network, you can converge the fixed and mobile architecture, you can converge the business and residential architecture.
You can also converge the layers in your technology, meaning you can collapse layer 3 to layer 1 using ZR2, still maintain the flexibility to select whichever optical vendor you want because of the disaggregation.
But now you have one box that does DWDM switching and routing instead of different boxes that are inflating the size of your sites.
And last but not least, you can converge your routing infrastructure.
You can reduce the number of IGP domains in your network if you are going for instance for SRV6, which allows you to simplify the routing architecture.
So convergence is the second major step in building a new network.
Automation
The last step is automation.
And I’m talking all kinds of automation, AI Ops or plain automation without AI.
At the end of the day, you are going a step forward towards some kind of autonomous network.
That means that you are reducing significantly the operational hub overhead and your workforce can focus on on generating new revenues and be much more efficient.
So this is what you as a service provider need to do with your network in order to survive and maintain a valid business.
3 things you need to remember.
The first step is disaggregation.
Second step is convergence.
What kind of convergence?
And the third step is automation.
Remember, disaggregation, convergence and automation.
Thank you, Dudy, very much for explaining that.
Thank you for watching.
See you next time on CloudNets.
Bye.