S&P Global’s Multi-Cloud Network Transformation Journey With Itential & Alkira

Chris Wade • 00:01

Hello, everyone. My name is Chris Wade, CTO with Itential. I’m joined today by Atif and Guru from Alkira and S&P Global. We’re here today talking about a partnership between Itential and Alkira we’re proud of, and we’re with our joint customer, Guru from S&P Global. So with that, maybe a quick introduction, Atif. Hi, everyone. It’s great to be here with Chris and Guru. I’m Atif Khan, CTO and co-founder of Alkira. Hello everyone. Great to be here again. This is Guru Prasad Ram Murthy, Head of Global Network and Infrastructure at S&P Global. Awesome. So we’ll get into it. So we’re super excited about this partnership. Maybe to get started, Atif, if you can give a quick introduction of who Alkira is. Sure, Chris. That’s a great question to start with. So I’ll just go briefly over why we started Alkira, when we started, and just a 30-second overview of what we do. So we started this company to address the challenges and complexities which were associated with networking in cloud-centric and hybrid IT environments.

Chris Wade • 01:22

So what we did, we built a global networking platform which is available on demand and consumed as a service. And we’ll go into more details as we proceed, as we go through this webinar. The main thing is it’s on-demand, global, and it’s consumed as a service, it’s available as a service, so I’ll provide more details. It provides a one-way or unified way to connect and manage networks across, I would say Clouds, across regions, and on-prem environments. At the same time, it ensures consistent security policies and compliance across these diverse environments. Obviously, networking in this day and age is not about just providing connectivity from one endpoint to another endpoint. You have to secure all of it.

Chris Wade • 02:14

There are many other networking and security services that you have to make sure that you offer and you support. We’ll go into some of those details as we go along. Awesome. And I’ll just give a quick Itential intro. So we’re excited to see technologies like this emerge. We created Itential with the thought that networking was going to be programmable nine years ago. So here we are.

Chris Wade • 02:39

We’re having networking turning into software. We’re going to talk about multi-cloud and that brings a whole scope of challenges. So we’ve had many years of innovation and networking and programmability is expected today. So we’re excited to kind of share how these solutions work together. But maybe first, Guru, you can give a little bit of background on the context for S&P Global and we’ll get into some of the challenges. Chris, thank you so much. I want to introduce the company and then we will introduce the use cases and the partnership that we are working with Itential and Alkera.

Chris Wade • 03:15

With that, I’ll get started, right? S&P Global is a leading provider of independent ratings, benchmarks, analytical and data for capital and commodity markets. We’re a $12 billion revenue company with 45,000 customers worldwide, working from 150 countries across the globe. To serve such a big customer landscape in 150 countries, we have 37,000 employees working from 80 different offices, leveraging thousands of our applications hosted in 50 on-prem locations, four different multi-cloud providers, providing that essential intelligence to keep the markets moving. Think about a day in the life for all of us. The markets has to keep moving. The financial services and financial markets needs intelligence in order to make decisions.

Chris Wade • 04:05

In order to make that decision with conviction, S&P Global provides that intelligence and data and the analytics and the benchmark needed for these many customers. And our employees work globally to empower these customers with necessary data. And an infrastructure behind the scenes that is empowering these many employees is significantly huge. And it’s multi-cloud. It’s a combination of on-prem. And it’s a combination of automation. And we have about over 100 agile teams working globally, serving these customers, releasing brand new features.

Chris Wade • 04:40

We have about over thousands of infrastructure specialists and security specialists working to help protect the brand and continue to keep the company at pace. It sounds pretty simple so maybe we can go through some of some of the challenges and where we think you know what was the typical enterprise infrastructure landscape look like. Well, that’s a great segue into what a large enterprise looked like. You take any large enterprise today, either it is S&P Global or somebody of our size or even bigger than that, what you’re seeing in the picture on the right side is something very unique, very common, and pretty much followed as a blueprint in any large enterprises. The reason that everybody got here is a combination of both organic architectures and inorganic adoption of our technologies, because as things proliferated, on-prem became cloud, cloud becomes SaaS, and the services become even more consumable. Every single technology department had different strategies to accelerate into the digital infrastructure landscape. It could be an identity stack, it could be a networking stack, or it could be a server and storage stack.

Chris Wade • 06:02

Everybody said we wanted to move into this digital infrastructure, which is a big landscape, but the adoption and the competition in the market is huge, big. But by doing different strategies, it completely shifted the user experience. As apps went into the cloud, the users connecting to the cloud is changing. As apps went into the SaaS, the users started changing. During the pandemic, the experience of the users even significantly shifted. The digital infrastructure of the past do not provide the capability to serve the user experience of today, and that’s a challenge that every enterprise is facing. To help those user experience, enterprises try to say, let’s do automation, let’s do DevOps, and let’s do Net DevOps, and DevSecOps, and whatnot.

Chris Wade • 06:49

And they built large DevOps teams to follow this infrastructure as a core concept, trying to tool chain every single open source component. It could be Ansible, Terraform, Jenkins, Chef, anything that could matter, but they try to build this high focus on infrastructure as a core concept. And if you see in large enterprises, you have big DevOps team trying to code this whole digital infrastructure. And the challenge is the maintenance of the code, the maintenance of the open source platform. And then, As things proliferated, as things become cloud and SaaS, the architectures become even more flexible and even more non-standardized, which led to the non-standardized hybrid cloud infrastructure integrations. And also, as things become complex, security team said, we have to start bolting security in every nook and corner as possible to protect the company, to protect the assets, to protect our people and brand and whatnot.

Chris Wade • 07:44

As you can see in the picture, there is a security for network, identity, server, storage, SaaS. There’s so many security products in the market out there, which is used as bolts and bells and whistles to basically protect this digital infrastructure. And this is leading into the cost, the complexity in the digital infrastructure, and the maintenance of automation is continues to be the ongoing challenge in delivering the speed and agility to our users. Atif, if you’re in the middle of this, what do you see from Guru and your other customers? Yeah. This is a great slide. We see exactly the same thing at majority of the large enterprises. As Guru is mentioning, services are now, they’re moved into the Cloud, they have SaaS.

Chris Wade • 08:34

They’re not just in one region, they’re spread across the globe. You could have some services in Europe and some services in Asia depending on the services, depending on the Clouds, depending on the data centers, like what’s running where. Bringing all of this together nowadays is the key. Simplifying all of this connectivity from the network perspective, making sure that it’s all secure, has become the key for these large enterprises. I would say the most important thing is the agility. I think that’s it for this session. If you have any questions, I’ll be happy to answer any questions that you might have.

Chris Wade • 09:11

Thank you so much for your time and I’ll see you in the next session. because nowadays like Guru and S&P or any of these larger organizations, they’re able to bring up these applications or these services like within a very short period of time. Since these are not limited to just the data centers, they can be anywhere across the globe. Now the network also needs to be agile and needs to be able to provide not just the connectivity, but the security to connect to those services so that you maintain the user experience because you cannot wait for the network for days or months for the network to provide connectivity when the service is up so quickly. So all of this needs to come together. So it’s putting a lot of pressure on the network side. And these large enterprises are looking at companies like us to solve those problems or provide platforms which they are able to leverage to address those challenges.

Chris Wade • 10:14

Yeah, sometimes we say, you know, our network exploded on the internet in the sense that we used to have a handful of data centers and a handful of WAN sites, and now you look at a drawing like this, you know, what is our network? At the end of the day, we have to connect application owners or users to their compute, and it could be distributed across a whole bunch of these areas. So I guess maybe Guru, who are the ultimate users of this infrastructure? Because from an Itential standpoint, we like to think that a lot of this infrastructure is traversing many, many boundaries, and if we’re going to offer same-day service to our end users, automation and orchestration is key. So you might have domain tools, as you said before, but ultimately we need to tie all this together. So who is the user that’s asking for changes in the infrastructure? And maybe something about what the velocity of those changes are.

Chris Wade • 11:11

Absolutely. Right at the beginning, when I introduced SMP Global, you heard I said we have about over hundreds of Scrum teams working hard in a safe and agile practices to deliver code in a CI-CD fashion at speed to our customers. As you all know, the software development world over the last 15 years have evolved to become a full CI CD with the speed at which that they can move, but it’s not the common in the networking world. While it has always been in the places where we have to catch up with them, but those business teams are our drivers, asking us, you have to become like a software world and treat infrastructure as a code services as a code, and then bring it as a as a service to on top of them. If I simply classify the types of users, I have a user who is a pure play developer who wants to see everything in code way, which means taking a service abstracting it as an API and giving it as a code to them so that they could add it to their pipeline. I have another type of user who wants to say, listen, I don’t want to waste time in your coding, you finish all your coding, just give it to me as a true blue white glove service so that I could just do full automation and then put it under a nice service, no platform or any other platform and give it to them as a service. The third person is basically the end user itself, who is doing the continuous CI CD services. So we have hundreds of scrum team working across the globe, trying to deploy code 24 hours in a window, the expectation is to this infrastructure should be agile enough to consume services that they’re asking around the clock across all multicloud providers on-premises and offices.

Chris Wade • 12:52

Yeah, and I think a theme I’m hearing, you know, you said NetDevOps earlier, you said DevSecOps. I made a mention about how things are kind of, you know, unconverging across the internet is, I think, whether it’s out here or Itential, we’re talking about how do we reconverge this, right? How do we build strong integrations so that the complexity in the infrastructure that’s needed by the business is not fully exposed to our end users, right? Because if the public clouds train them on anything, they want self-service, they want ease of use, they want consumability. So I think, you know, for the definition of the problem in the business, we need to take advantage of all these things and modern networking infrastructure, software, hardware should be extremely programmable so we can apply these operational paradigms and ultimately offer a great service to our end users, whether they be application owners or end users. Maybe we can talk a little bit about the partnership that supports this, but maybe a little bit of the definition of the problem, Guru, as you see it. Sure. Just to piggyback on the previous slide, think about an enterprise in that state, and if you simplify that large digital infrastructure into small circles and start visualizing this both from a problem statement, as well as from a business case perspective, why we are doing this, why is this partnership and integration is needed. Think about the digital infrastructure picture that we just showed, how many wires and how many connectivities and how many configurations have to go in place in order to enable all the upstream customers and the downstream services. When I say upstream customers, to deliver services to our business and business teams, to have the agility and speed and our downstream services, such as it could be SOC or information security or workplace services in order for them to consume our services. Network is in the middle and we are responsible for providing services to our upstream customer and our downstream teams. So doing an integration on an unconverged network is the biggest problem statement and that’s something that every large enterprise are trying to tackle today. You have a firewall, you have load balancers, you have routers, switches, cloud, and all these are available in cloud as well as cloud native services. So you want your developer to consume a load balancer in the on-prem, a similar load balancer available as a service inside the hyperscaler as well. So they’re expecting, I want both of these to converge. So having that convergence and creating that convergence is the biggest challenge and we continue to do that effectively and that’s what we are trying to solve. Now coming to automation, we talked about

Chris Wade • 15:42

how large enterprises are setting up this large scale DevOps teams and trying to build every single component that we have mentioned in the screen by themselves. It could be from writing a workflow, writing a user portal, writing a rules engine. Building a full CI-CD pipeline is a factory worth of exercise, right? And you need to have skill sets and talent and ability to maintain those technology and the code and the services throughout the lifecycle to build a full suite of automation. What at S&P Global we are trying to translate all of this problem through the partnership of Alkira and Itential by bringing two large universal solutions that can take us from an unconverged network to a fully converged network using Alkira and doing open-source automation to a full low-code no-code automation using Itential and it takes time to get there and we are already there and we have had years worth of partnership with both Alkira and Itential and and for every enterprise out there, when you take this transformation journey, give the time to get to this transformation so that you can adopt some of these universal platforms. Exactly, and I’ll just add a little bit to it also, you know, From the traffic or from the data perspective, now, as we mentioned earlier also that your data is everywhere, it’s not just in the data centers, you have apps running everywhere.

Chris Wade • 17:15

Now, from a network perspective, agility is one thing, but also you need to optimize the traffic flows as well. Because your users are everywhere also, and they’re not necessarily just sitting in one building or just few buildings, they’re coming from everywhere. Your traffic flows, you cannot take a non-optimal path from that perspective or backhaul the traffic to data center and then to the Cloud where the application is running. That adds a lot of latency and stuff, and that directly impacts the user experience. From a network perspective, network has to be agile enough, network has to be flexible enough that it takes into consideration all of those things and make sure that the user experience is not impacted and you get the lowest latency path to your destination. All in all, I think networking has changed or needs to change from a deployment perspective going forward because just deploying the way we used to deploy networks doesn’t cut it. And as we mentioned, all of this needs to happen like now, like when somebody needs a network to accommodate their needs, it needs to happen right away.

Chris Wade • 18:40

It cannot just be that, oh, I have to order like circuits or I have to order hardware to make it happen. So all in all, I would say it’s all changing, it’s all changed, and the needs from the requirements from the customers are very different today as compared to what they were a few years ago. Yeah, maybe I’ll start with a compliment in the sense that, you know, on the right-hand side here, we have Alkera plus Itential, but. So, we talk about it a lot in the sense that we expect modern networking solutions to come with programmable APIs, and from an Itential standpoint, we work very hard so we can consume those APIs. We expect two modern platforms to work well together. We don’t expect Guru to be hiring SIs to integrate our two tools, right? So, the integration took minutes and hours.

Chris Wade • 19:31

You know, we work together on what those joint kind of pre-builds or solutions should look like, but the expectation here is, and that’s maybe a little bit of hope here in the sense that when I look at a lot of the open-source automation, all automation is great. A lot of times, we automate for ourselves. You know, it takes us a long time to do stuff, so we automate that, but integrating these automations together to drive that end-to-end process and connectivity from app to user to application has sometimes been a challenge, so we’ve been fairly limited in siloed in our automation effects. So, really taking some of these modern networking concepts and applying kind of modern orchestration with automation really allows us to have and provide our end-users, you know, robust, real-time, consumable infrastructure, so, you know, I think it’s probably worth noting it wasn’t like a huge effort between us, but we are super proud of it, and at the end of the day, the effect here is, and I think it’s one that the market should expect and drive towards, is that when they consume modern networks and they look at orchestration and automation, these things should work, you know, more out of the box, and we should be focused on the business rules and outcomes and not focused on, you know, integration like we have in the past. I’ll just add one more point right the out-of-the-box capability and the canvas that Both Alkira and Edential has built is a game-changer to us Think about going and managing hundreds of VPCs in a cloud native AWS console To what we do in Alkira in a single canvas. We can clearly visualize how the network is Orchestrated how it is connected and when to do what conditions and think about trying to automate a network and today At S&P Global with Edential partnership, we’re not writing code We have the canvas and all we are doing is pre-integrated conditions are joined together Based on our needs and that canvas is driving the outcome of it Which Chris and team are heavy lifting in the back end to bring the pre-integrated Capabilities to S&P Global and we are able to put all of them together and deliver it as a full service to our customers Now, that’s great to hear, Guru, and thanks for sharing this. And I’ll just add a little bit to it also that from the management perspective and then from the operations perspective, you know, networks, there’s stuff that goes wrong every now and then, and it’s bound to happen, it’s software, like any software will have issues.

Chris Wade • 22:10

So you need like full capabilities from troubleshooting and how to do all your day two operations. All of those need to be built into these environments. Because troubleshooting, when something goes wrong, how quickly you’re able to identify the problem and how quickly you’re able to fix the problem, I think that is the key. And those have improved in recent years in networking products. And we take pride in our platform also that we built this platform with day two operations as one of the key focuses of ours. makes a lot of sense. So maybe we can get into a little bit more of the specifics as far as like, you know, when you have branches and data centers, and you’re trying to connect to a multi cloud environment, like, how does that work guru? Sure. I would like to do a little bit of recap, right? We talked about

Chris Wade • 23:15

the digital infrastructure landscape of where the enterprises are today with hundreds and thousands of connectivity, different infrastructure components, different security components, everything is tightly bolted and the infrastructure is constantly proliferating and it’s evolving constantly, right? And then we talked about how the integrations have to be done both within the infrastructure side and within the automation side to glue all of them and this is the meat of the whole story, right? How we solved this complexity and challenges at S&P Global, right? I’ll start with a simple analogy. Alkira is a super glue for us in the middle of a large hybrid network. Like I described before, we have about 80 branch officers, 50 on-prem locations, and then four multicloud providers over thousands and thousands of applications in AWS, Google Cloud, and Microsoft. Think about doing something like this at scale in the old architecture way that we just showed is going to even more make the environment complex and challenging. So we brought that superglue technology to basically stick everything together and then fully abstract the entire network, both from the branch, from the data center, from all the three cloud providers and put it on Alkera. So on their canvas, we were able to manage all the routing and security requirements and clearly define which one has to go where and optimally define the best path that it can go.

Chris Wade • 24:59

So one of the best things that we accomplished is historically our applications teams have challenged us, hey, my application have gone to the cloud, from on-prem to the cloud, which means I’ve transitioned from a physical on-prem server to a nice EC2 server, which means my availability of my application in the cloud is as equal as what AWS gives us as availability. The challenge the network team faced is when do you think your availability will match AWS availability? Here we are today, we have taken our availability by adopting Alkera and because they can run in AWS, Azure and Google, we were able to match the availability of this hyperscalers into the network. So if the application is running, my network is running. If the application is down, my network is equally down. There is no, there’s not a day that my service is down, the app is up and the app is up and vice versa. So we have matched that availability and the expectations of our business by taking our availability to a hyperscaler availability. Now when we did all of this, our business said this is great, you were able to improve the latency, you were able to reduce the round-trip time and what not happened and everything came together. Now the worry from the business is, Guru, what will happen to the agility? Suddenly you’re introducing a new technology, none of us knew about what is this, but you have a super glue in your hand where you can just stick everything together like it. But what will happen to my past automation that I have built inside AWS, inside Azure, inside GCP, inside VMware, how will I handle all of this? That’s where we brought this heavy duty rubber band around it and then we said, you were consuming all of this in the past through iTensile. You did not know about it, all I gave you is that nice APIs. through ServiceNow platform, you are able to consume all of it. Now, we just expanded that rubber band around this entire landscape. And with the partnership, what Alkira and iTenshal brought together, automating the VPC onboarding, TGW onboarding, security group management, all of these capabilities are built within this heavy-duty rubber band. and every component that we discussed within the digital infrastructure landscape, a branch office will have a router switch firewall, a data center will have router switch firewall load balancers and whatnot, and then you can assume the scale of the cloud.

Chris Wade • 27:25

So ability to automate the entire network services on this such a large landscape requires a heavy-duty rubber band around it to deliver this as a self-service, which gave a great confidence for our business. Now I understand how are you connecting the glue and the rubber band and holding this as a service, which is much more reliable and much more consistent for us. And by doing all of this, what we accomplished is we were able to deliver a new region completely from the ground up in less than two days, which used to take two months, and we came down from two months to two days, delivering a completely a new region. We were able to integrate some of the existing SD-WAN solutions that were implemented prior pandemic because SD-WAN was popular at that time and every enterprise today have one of the other SD-WAN solutions. So we were able to naturally bring the SD-WAN solutions into the Alkira ecosystem and keep the automation punched through the Itential system. This way we were able to consistently measure the speed and agility with the business. Thank you.

Chris Wade • 28:31

Then, when we are re-architecting this at scale, our security team said, how are you going to enforce policies? Because I built my SCCM policy in AWS. I have another Terraform policy in Azure, because this is where I built before all these capabilities were in place. So we were able to demonstrate that by building that uniform standards, uniform compliance through Itential, and enforce it consistently in every single component of the digital infrastructure. And on the right side, what you see is how we got it done. And today, at S&P Global, think about enterprises of our scale where we used to have, back in days, big Verizon-like WAN, AT&T-like WAN. Now, today, we have the ability to replace some of these big, large service providers into enterprise exchanges of Alkeras that runs on a hyperscaler, and then eliminate all the one-off automation into an Itential canvas-based automation so that we can keep up with the speed and agility of our business.

Chris Wade • 29:36

So maybe one question on the how, because we’ve all provisioned a lot of VPCs, Transit Gateways, VNets, that kind of stuff. How does this work? How do I coordinate the connectivity between maybe my applications and my regions with these Internet exchanges? I think that’s a great question, Chris. Thanks for asking it. I’ll go into some details. What we have done is we have built our own infrastructure, which sits inside the clouds. As Guru mentioned, we leverage AWS, we leverage Azure, we leverage Google Cloud Platform.

Chris Wade • 30:11

So our infrastructure is available anywhere where there’s a cloud region. So it’s globally available and it’s a multi-tenant infrastructure. Now, let’s say a customer like S&P Global, they come and they can start building their network on top using our infrastructure which is always available. Guru mentioned earlier that availability is a very, very important factor for companies or large enterprises because the Cloud is pretty much always up. I shouldn’t say always up, but it’s as available as it could be for right now. If the Cloud is up, then the network needs to be up also. But we built this infrastructure in such a way that it’s highly available within Cloud and across Cloud.

Chris Wade • 31:08

We provide disaster recovery across Clouds as well from the network infrastructure perspective. We provide horizontal scalability from the infrastructure perspective. This infrastructure is not just a router sitting in there or something, it’s our own cluster of nodes. It’s a scalable infrastructure. We keep adding more nodes to add more capacity to the infrastructure, and we onboard customer’s network on top of it. That’s how customers are able to leverage the infrastructure to quickly build those networks. They build their virtual constructs on top, and we call that virtual construct Cloud Exchange Point.

Chris Wade • 31:51

Each customer comes and builds a Cloud Exchange Point leveraging that infrastructure. And now from the cloud exchange point, they can connect any location, on-prem location to it. They can connect any cloud to it. And they can connect any SaaS application to it. That’s the connectivity part, but then there are capabilities of running multiple network and security services on that platform as well. If you look at these large enterprises, S&P is one of them, they have standardized their security on some large security vendors. They want to do security in the Cloud the same way they’ve been doing security or running the appliances on-prem as well, and manage those firewalls just the same way.

Chris Wade • 32:40

We allow them to instantiate or run those security appliances inside of this platform. In fact, what we have done is we have gone an extra step where we optimize the deployment of these services inside of this platform. Before, let’s say if they had to deploy X number of firewalls to meet their use cases or requirements, they can deploy many less number of firewalls because we optimize the way we deploy them, we optimize the way the traffic is steered through these firewalls. They are able to save a lot on the capacity that they need from these services also. Then this infrastructure is global, as I mentioned. They can spin up a CXP, which you can think of it as a virtual colo or a virtual pop, with all the infrastructure already available to them. They can bring it up anywhere across the globe.

Chris Wade • 33:37

As Guru mentioned, they had to bring up a region, and they instantly brought up this region, which they had to. Since it’s a networking platform, so when we built this platform, our focus was that, obviously, it’s sitting inside the Cloud. Multi-Cloud networking is one of the biggest use cases for it. but it’s a networking platform, so we have to enable this platform to address other use cases as well, so networking use cases. We have been able to, just because we took a platform approach, we’re able to address those use cases as well, and I’ll give you examples of partner connectivity, Extranet as a service. where your partners are coming into this platform. They can be global, they can be coming from on-prem locations, they could be coming from the Cloud side, and consume services from S&P’s network, whether those services are sitting in their data centers or whether they’re sitting in their Cloud, we can selectively do all of that stuff.

Chris Wade • 34:38

Then it’s a fully routed platform which is global, so it also provides full SLA-backed high-speed connectivity, fully routed, fully segmented, full mesh connectivity across the globe as well. Some of our customers are leveraging this platform, the global platform for the global connectivity as well, high-speed connectivity. There are multiple use cases which we have been able to address just because we took the platform approach. We connect to all these Clouds in one way. From a customer’s perspective, they just come in and they say that I need like X gig of bandwidth to this VPC or to this VNet, and we make it happen for the customers. We use the underlying construct of ours, underlying constructs of the Cloud, but it’s one way of doing networking for these customers. Obviously, there’s a management and policy layer on top of it, which helps them manage the network, run the network, day-to-day operations, troubleshooting, all of that stuff.

Chris Wade • 35:48

Awesome. Maybe a little content pitch. I think our teams have recorded a few demos showing iTential and Alkira integrated together, and as you said, also integrating with third-party security services with their own management console and building some of those end-to-end services, so all super exciting stuff. Yeah. Bringing all this together with the iTential just makes this platform much more powerful because now for customers, it’s one way of doing end-to-end automation. Alkira is part of it. Automation, as we’ve been discussing, and as we know, it has become the key. Ops, network operations have moved to DevOps approaches. That is the key. We are very happy to be partnering with iTential and I’m glad we’re able to help customers like Guru along the way.

Chris Wade • 36:42

I’ll probably add a little bit of context to support the initiative that we are doing. S&P Global is in this multi-year. cloud transformation journey like a discrete we have about 50 on-prem locations. Think about you’re a developer and then you’re empowered to choose a provider of your choice, right? You’re a developer who built this platform you might know your app might work in Azure or app might work in AWS or app might work in Google, right? Giving that 12-lane highway in order for the developer to take their app where they intend to take and and deliver the best outcome is the most powerful use case we were able to accomplish and that is through the Alkira and Itential partnership. That’s that’s number one. We have about hundreds of apps that are yet to be migrated into the cloud and our scrum teams and development teams are actively working on moving to this cloud and that flexible autoscale highway gives them an ability to move fast. Sometimes we come into a conversation with teams who are migrating petabytes of data saying, listen, we need more bandwidth for the next couple of weeks. I cannot go and tell them in this time and age, give me 90 days, let me order a circuit from somebody else and give it to you. With the ability to run Alkera inside the AWS, and AWS is a hyperscaler, we were able to provision the ingress and egress to the maximum capacity that we need.

Chris Wade • 38:18

And this highway is fully automated and we were able to do that. And kudos to both of you because the technology that both Alkera and Itential built, that agentless architecture, is such a powerful use case. Today, you take any infrastructure. components out there in the market, most of the security components out there in the market, you need to run some sort of agent in the network in order for them to collect and connect and communicate. The agentless architecture of both Alkera and Itential makes our life even more easier because so that I don’t need to worry about I have to connect my mother ship to another child ship, child ship to mother ship and then you become another on-prem hybrid ecosystem. Taking this to fully at a canvas level and abstracting it is the power of both the technologies that we were able to take advantage in our implementation. Awesome. Maybe Atif, a bit on the company, I know we’ve spoken more about the specific challenges, but anything you would like to add? Yeah, definitely. A little bit more to it.

Chris Wade • 39:26

Many of our users or customers, they have multi-Cloud. Not just multi-Cloud, but these multi-Clouds are in multiple regions as well. You can think of it like if they’re in four Clouds, it’s four Clouds and multiple regions of each Cloud. Now all of this needs to come together. The way we bring all of this together, this diagram basically just gives a high-level overview of how we do it. As I mentioned, it’s our infrastructure which is available across the globe. These CXPs are the virtual constructs which each customer is able to bring up anywhere using that infrastructure across the globe.

Chris Wade • 40:11

Then they’re able to connect to any Clouds from any of these locations, any on-prem location or user to any of these CXPs as well. The moment you connect to these CXPs, you have any endpoint to any endpoint connectivity. Just the use cases which we are enabling, obviously on-prem to Cloud, inside the Cloud. There’s a lot of networking that needs to be done inside the Cloud as well. You have to take into consideration a lot of these things, intricacies of each Cloud, which is different in every Cloud. We can go into those details at some point. Cloud to Cloud is a big requirement.

Chris Wade • 40:55

In fact, Guru started with this Cloud to Cloud use case in the beginning between GCP and AWS. There’s a lot of Cloud to Cloud requirement, and then you need to insert security services in the middle of all of that stuff. Then as I mentioned, it’s one way of doing networking across these Clouds, one way of securing all these Clouds, and then optimizing the deployment of these services from security services to other networking services, and management. Management is a key as I keep saying that, because day-two operations, you deploy a network once, but you’re managing this network every day. You have to have visibility. You cannot have blind spots. Capabilities where you have all the visibility in your network, from the flows perspective, traffic perspective, other metrics and latency perspective, where are you having drops and stuff.

Chris Wade • 42:05

Those capabilities are as important as anything else. One last thing I’ll mention is consumability. we see that networking is moving towards as-a-service model because it has to move to as-a-service model because it needs to be agile. It needs to be there when you need to connect somewhere or you need to secure something. So that agility, that flexibility, that large footprint where you’re able to extend this network anywhere across the globe without you having to buy hardware or build it yourself, you should be able to just buy it, consume it and secure it. It’s still your network, but you don’t have to worry about the infrastructure. Just like when you bring up a server in EC2 environment, you don’t worry about where the physical server is running.

Chris Wade • 43:05

So same thing has to happen with networking where you shouldn’t have to worry about the infrastructure. So that’s what we have done. I call it network infrastructure as-a-service because it’s our infrastructure as-a-service, but it’s your network which is running on top of it. Just to double-click on a couple of powerful use cases. Historically, to connect this hybrid infrastructure, network teams of the enterprises used to build this physical colos in order to connect a cloud with an on-prem or a branch with a cloud and whatnot. And that colo essentially have become a central heart or a life system of keeping everything together. What is a unique thing that we’re doing with Alkera is about transforming the physical colo into a carrier neutral colos, which runs in the hyperscaler so that I don’t need to keep the lights on in the colo using physical devices.

Chris Wade • 44:02

As physical devices, as you run, the availability is a challenge. You have to keep a backup hardware. You have to wait for RMA to come. You have to do a code upgrade and whatnot because we are able to transform the physical colos into the virtual carrier neutral facilities in the cloud and running it on the Alkera, we were able to use a super glue and attach all the hybrid infrastructure in the cloud together. That’s one powerful use cases. S&P Global is a data company. At the beginning, we talked about 45,000 customers in 150 countries worldwide.

Chris Wade • 44:35

And we are in the business of distributing data with our clients and exchanging data with the clients. And with those many clients, it is not practically possible to run a physical wire to their infrastructure. And because as a client, they are also transforming. As a business, we are also transforming. And Alkera is becoming like a snowflake is a place where everybody exchange data. Alkera is becoming like a snowflake of the networking where they are able to bring the clients, the third parties, and the partner on the platform in a secure way so that we were able to eliminate this classic partner net infrastructures and then move them on a secure Alkera partner net infrastructure, which still stays in the cloud, but fully separated from the core of the network so that we feel secure, the customer feels secure. And we are still able to distribute data at a cloud speed.

Chris Wade • 45:33

Awesome. So maybe just a little bit about itential in the sense that, you know, we just talked about Alkira’s, you know, consumable, cloud-centric networking. My words, not Atif’s. So, you know, the expectation there is that it comes with automation, I think, is the point, right? You’re not providing boxes and virtual machines here. You’re providing automation within your infrastructure. And that’s what we expect from modern technologies. So, you know, from an itential standpoint, when we talk about automating infrastructure, a lot of times we’re automating automation. And a lot of times we’re automating automation that’s embedded with, you know, some sort of software or equipment provider. And that’s really where the industry is going. So when you look at the automation layer within itential, we have this adapters integration concept, which allows us to do real-time integration with open API spec platforms like Alkira. And then we also have our automation gateway, which talks to devices that are programmable like GNMI, NetConf, et cetera, but we also integrate and treat Ansible, Python, Terraform as first-class citizens within the automation gateway. So across that integrated ecosystem, we can talk to all of these things, security platforms, SD-WAN platforms, cloud platforms, et cetera. So the first key to a broad-based automation solution, like we’re talking about, is to be able to integrate with everything and take advantage of all the automation that’s coming. We’ve moved away from the vertically integrated tool-centric world to a world of platforms, and having an automation and orchestration platform to tie these things together we think is quite critical. So, after we have that automation in place,

Chris Wade • 47:08

It’s really about, as Guru mentioned, the low-code canvas, but tying these things together and making it very simple for more people to participate in automation. We can’t have all the tickets go to a group of people in the corner to do all the automation for the whole business. Automation is how we operate the network. We think it’s the operational paradigm of networking moving forward, so really tying all these things together under an orchestrated umbrella is very important. But at the same time, we also have to make sure we’re doing validation and we’re adhering to the business rules that we’ve put in there. Automation Studio is how we build automations, Config Manager is how we put validation in place to make sure we have some guardrails to our automation, and ultimately, what you’ve heard a lot from us lately is how do we self-serve this? More and more customers are asking, how do I have my end user actually request automation rather than me swivel-chairing and doing it myself?

Chris Wade • 48:06

We’re moving from a world where automation is built for me to do my job better, for me to build automation for others to take advantage of. We’re usually asking people, who is the ultimate user of this automation? Because we need to meet them where they’re at, so when we think of our primary consumers of automation and orchestration, it’s really divided in two in the enterprise market. The first is platforms in the IT space, so it could be a ServiceNow, that’s what we talked about today. We’ve even published an app in there to make it super easy. There’s also Salesforce. A lot of times, if network services are viewed as part of the CRM, and that’s one side, and the other side is really line-of-business apps.

Chris Wade • 48:49

So we kind of IT, quote unquote, people requesting infrastructure, it might be day two, it might be maintenance activities, or it might be line of business owners, and they generally, you know, application teams, you know, generally want to talk in the realm of APIs and pipelines. So we need to really expose these automation and orchestration. If we build an automation and nobody uses it, it’s only of limited use, we’re using it for ourselves. But we’re highly interested in how do we expose these automations for our customers to use, So they can view, you know, our operational paradigm similar to what they’ve been trained with SAS providers and public cloud personas. Chris, maybe a question here, right? Network team at S&P Global, we serve about six business unit or thousands of developers in a day. A software development world have significantly advanced in the CICD world.

Chris Wade • 49:42

And today, the network and network automations is equally advanced. And we have a chair in the room to compete against them to provide them the service. What’s new in the room is, hey, there’s something called Backstage from Spotify. And that’s going to be the new developer enablement platform that we would like to consume. And then the ask from them is, how can you integrate or interoperate to make the developer experience not just consume network service in a separate compartment, but they would like to consume middleware, storage, server, all the code things into the thing. So if you can provide that IDP as an experience, that’ll be useful. Yeah, so where we think we’re going with the industry is one of platforms and a lot of people put it under the platform engineering umbrella.

Chris Wade • 50:35

The challenge as it stated, because a lot of times, especially even on this call, we talk about, you know, we’re talking about infrastructure out, right? We’re really thinking about infrastructure out. If you’re thinking application in, and you’re an application developer, sometimes you’re like, I’m not sure what I’m supposed to use, how I’m supposed to use it, what am I allowed to do? You know, line of business teams for a long time swept credit cards, you know, used cloud and other services as they saw fit. Most CIOs are saying, you know, we want you to use all those great services, but we would like to standardize those a little bit. We’d like to have a tailored, you know, S&P global opinionated set of services versus just whatever’s out there, right? So that, you know, people are starting to assign product managers to their platform teams. So you can look at connectivity as a product you’re offering your application owners. And ultimately, we need to talk to application developers in the manner of which they’re used to. So IDP, Internal Development Platform, is really the thought where application teams can go look at what’s available, kind of think of it like a console.

Chris Wade • 51:38

And from an IT central standpoint, we want to publish our APIs. You can publish them as you see fit. You can customize the routes so we can publish that to the application team so they can understand how to consume this network. And also there’s a level of abstraction in there so that you can change, you know, you’re talking about constantly changing some of your infrastructure, but we don’t want that to ripple up to our application team. So providing those as APIs really allows us to be innovative at the infrastructure level, but also communicate to application teams as they see fit. And, you know, I think the biggest concern slash issue they see is that they spend a lot of time trying to figure out what they’re allowed to do, how they’re allowed to do it. And you don’t want to get to build time and have the wrong the wrong set of services.

Chris Wade • 52:20

So us, you know, publish these in a more platform engineering concept is going to be a huge kind of leap forward as an industry. And I think that’s that’s that’s what we’re a lot of us are converging on. So excited to continue to talk about that for sure. I think that’s that’s very well said. Just read a couple of perspectives there on the on the item show. Many years ago when we first started. designing the strategy around DevOps and

Chris Wade • 53:03

bring in the full suit of network automation inside Itential because they today have integrations with over 100 different networking technologies and we were able to bolt all of them at rapid speed I think the vision that we carry forwarded with the Itential team and we wrote a great white paper together explaining it to S&P Global business and business CTOs What this means to them right and they appreciated a lot about the vision of low code no code two three years ago and we were able to produce that output in in real time today and second second biggest thing is one of the very recent use cases that we delivered in partnership with Itential is Today the scale at which the digital infrastructure is expanding and now we’re introducing Al-Qaeda and many other thing security wanted to Handle security events and bad actors instantly. We were able to develop or abstract The SOAR automation out of a SOAR platform and bring that intelligence on top of Itential platform So that today any SOAR platform in the market they are just providing a payload to Itential and Itential is able to Translate that payload a payload could be a domain name or payload could be a bad IP addresses so they were able to absorb the payload from any SOAR platform such as Phantom or D3 or anything out there in the market and translate that one payload into hundred different understandable variables and pass it to different technologies within the digital infrastructure. They were able to take the payload and process it on a proxy, process it on a firewall, process it on an info blocks, on Akamai, whatnot technologies within the digital infrastructure. Itential is able to translate the payload and apply it instantly. Today we have a capability that if you give us an IOC IP address we will be able to use Itential and block the IP addresses within the network in a matter of seconds and we do that actively. This way we enable and empower our front line cyber defense SOC team to action it immediately instead of putting a ticket and waiting for somebody in the network team to act upon it. That’s some of those powerful use cases we are working in partnership with Itential. That’s awesome. I appreciate it. It’s a great use case. Maybe just to explain a little bit behind that is the sense that we talk about infrastructure being programmable. As a company who tries to automate and orchestrate infrastructure at scale every day, that’s how we think. There’s been different attempts to standardize APIs and standardize payloads. Take SD-WAN, so much innovation so quickly, no standards. As we get into advice for those that have hung on with us, I would say that what we’ve learned in these use cases, like you were just talking about, integrating lots of systems, is that people don’t agree on payloads. People don’t agree how to describe interfaces. I mean, look at the public cloud providers, VPCs and VNets, different APIs, different payloads. But everybody uses them, right?

Chris Wade • 56:09

Because they’re well-described. They’re well-described with payloads. From an intentional standpoint, it’s not about trying to get everybody to agree on a handful of constructs, because we think innovation cycles are accelerating, right? We’re going to see more software-based infrastructure. Cloud providers are launching hundreds of services every year. Folks like Alkira are innovating at a huge clip. So what we need is we need those APIs to be programmable and described which which makes it automate, you know so we can automate this and Since we’re not going to agree on payloads. We have to be able to do data transformation at runtime And guru, that’s what you’re describing, right? So we can understand how info blocks or somebody else describes the interface and then we can automatically do Schema transformation on that. So there’s all sorts of great technologies Available for modern infrastructure and I think you know Like I said before one of hope here is that folks that are running scripts and doing some some basic automation all automation is great But when we start to put multiple systems together in multiple domains You know, there’s some additional techniques that allow us to really do this and it’s available today But maybe as we get into advice for those that are still here with us, maybe talk about some of the things you’ve seen You So we see that some of our customers are predominantly in one cloud, but they are leveraging other clouds now for different reasons. And it’s actually the business that drives the adoption of the cloud because there are certain business needs which drive that. So I always say that if you’re in just one cloud, you’re one use case away from becoming multi-cloud or one acquisition away in some cases to become multi-cloud. So planning for multiple multi-cloud types of environments or deployments has become like a must for large enterprises.

Chris Wade • 58:21

And the other thing we are seeing is that, as we say on this slide, on-demand networks available as a service consumption models, networks as a service, which are available, which are agile, which are flexible and available on-demand across the globe. So that is the key going forward. And that is what we are seeing with large enterprises, mid-size enterprises, in fact, across the board. Awesome. And Guru, with some of your peer group listening, what sort of advice do you hand out? One of the key thing is when we decide to transform or when we decide to innovate, in the past, the mindset used to be, let me first take everything to the cloud and I’ll automate later or I’ll automate first and then I’ll take it to the cloud. You cannot gamify this. You have to transform both at the same time, at a rapid speed, because both of them yields to each other. If you can transform to the cloud, but if you cannot automate, you will lose security and agility. If you automated everything, but if you don’t go to the cloud, you will have the other challenges of the business as well. So the.

Chris Wade • 59:39

At S&P Global, we were able to see both sides of the coin at the same time. Going to the cloud is a business driver. Delivering agility is our internal driver. And we were able to take both sides of the coin, put the right tools and right capabilities around it, and take that innovation to the next stage. That’s number one. Number two is, in the past, either partners like yourselves waited for enterprises to come with problem statements, or enterprises like us waited for partners like you to come with world-class solutions. I think we are today living in a world where both enterprises and the partners have to work together, try to solve the problems together, and continue to co-innovate.

Chris Wade • 01:00:25

That accelerates innovation. That accelerates solving more and more use cases. That brings in modern capability to the market so that we can take the network to a completely next level. Yeah, and as you guys said, the common theme here is thinking about our end consumer, this application developer or this end user. They want the cloud that they want, like Atif said. We have to orchestrate this together, and there’s no operational model that’s going to meet them where they’re at without aggressive automation and orchestration. So I’ll end there.

Chris Wade • 01:01:02

I wanted to thank you both for joining us today, Atif and Guru. Thank you very much to those that are on the call. Thank you for attending, and we’ll see you next time. Thanks, guys.