Top Emerging Cloud-Native Trends | Cloud Native Now 2023

Hello, everyone. Uh, very good morning, good afternoon, and good evening wherever we are located. Uh, welcome to one another, uh, cloud native Now session.

Uh, in this, uh, session, we are gonna discuss about some of the, uh, top emerging, uh, cloud native trends. Um, I'm Ganesh Kumar, joining you from India. Um, so I work in Tata Consultancy Services, also known as, uh, tcs.

So for those who may not know, uh, data Consultancy Services is an IT services consulting and business solutions organization that has been enabling customers and their transformation journey for over 55, uh, plus years. So, within tcs, I lead the container and DevOps practice team. Uh, that includes, uh, Kubernetes and cloud native ecosystem, uh, solutions and offerings in Agile, uh, infrastructure center of excellence.

Um, uh, this is an infrastructure practice group that comes under a large, uh, infrastructure services, uh, unit called Cognitive Business Operations. So, coming from, uh, you know, uh, the Center of Excellence group and observing the industry trends across, uh, several industry verticals, we are already witnessing, you know, the cloud native technologies becoming more, uh, pervasive in many, uh, industries, um, that makes it, the NextGen, uh, enterprise platform will definitely hold containers and Kubernetes, uh, as its, uh, native, uh, citizens. So part of, you know, uh, digital, uh, transformation initiatives.

Uh, we also see several enterprises are at, uh, various, uh, stages of their cloud native, uh, journey. So, uh, large, uh, global organizations are also in investing, in building, uh, communities of practices to, uh, keep innovating continuously, uh, especially towards their, you know, hybrid and multi-cloud, uh, strategies. So, with that, um, uh, I'll jump on to the today's agenda, what we are gonna discuss, right?

So overall, um, we observe, uh, that containers and Kubernetes adoption is, uh, constantly increasing year on year. And, uh, Kubernetes has already emerged as the defacto platform for container orchestration. So it is also evident, right, uh, that many enterprises are, uh, considering containers and microservices, uh, for their modern, uh, software delivery.

So this has led to the discussions on what's next on this, uh, cloud native world, and what more can be derived out of it. So this new, uh, paradigm has, uh, opened new operating models and new models of, uh, consuming the Kubernetes platform so that, uh, you can extract more, uh, efficiency and productivity out of it. So, with that, uh, you know, the, some of the, uh, top trends that, uh, you know, we will, uh, discuss today are, uh, vm uh, the virtual machines and, uh, containers, uh, convergence, and then followed by, uh, running state full workloads, um, along with, you know, data protection, um, and followed by, uh, cloud native security and cloud native, the platform resiliency.

So these are the four, uh, you know, uh, topics that we would discuss. Now, of course, uh, while these topics, you know, uh, it may not sound, uh, something new in this, uh, cloud native, uh, world, but, um, as these have been, you know, already prevailing for, you know, some time now, but the, uh, what we see or what we observe is that it's apparently complex to productionize, uh, these, uh, technologies and especially when you operationalize, uh, at the scale, right? So it becomes, uh, we still see enterprises are struggling and exploring the solutions.

So that's why, uh, uh, we have picked up the, some of these, uh, topics to discuss today. So, with that, um, let us pick up these topics one by one and, uh, discuss more in detail. So, to start with, you know, um, looking back, uh, the primary reason for containers and Kubernetes to originate.

Um, so it is primarily to convert or decouple the tightly, uh, integrated, monolithic applications, uh, into a lightweight and modular microservices, so that they become more easy to build or operate and can be, uh, you know, quickly scalable and also portable. So this makes, you know, life easy for the developers and operators. So this is where we started.

Um, so having said that, not all applications, uh, can be containerized in the, uh, day one itself, and that arises the need to run and manage, uh, both the traditional platforms like virtual machines, uh, for an extended, uh, period. Uh, for example, uh, when you have a partially modernized, uh, uh, an application wherein, uh, some portion or some parts of the application is still, uh, running on a legacy databases running on end of end of life, uh, operating system, uh, so we'll, it'll still exist on a virtual machine. So you may have to continue in this hybrid model until the application is fully, uh, refactored or retired, whatever the you apply, uh, on that.

So, thereby the need arises to co-locate these traditional, uh, workloads along with modernized, uh, container, uh, uh, containers, especially when it is, uh, you know, latency, uh, uh, centric workload. So this is one example. The, another, uh, example could be, uh, let's say building a private cloud, uh, set up, uh, using a, a cloud native virtualization.

Um, let it be to host and manage both, uh, virtual machines and containers in the, uh, on a same, uh, unified, uh, platform. So on these use cases, what we discussed, uh, we see that it becomes a mandate and also significant to run and manage both Williams and containers on the same platform. So this is where we see, uh, uh, the open source projects such as, uh, cube work, uh, which is becoming, uh, prominent and reliable, especially.

Uh, one other use case could be in remote edge, uh, locations where you have, um, you know, limitations on the infrastructure capacity. And, uh, within a couple of few, uh, available physical, uh, servers, we will have to host, uh, both VMs and container workloads. So where, uh, it becomes another use case, prominent use case where you can consider, uh, know, uh, managing both virtual machines and containers.

So the, this is one aspect of the technology use cases that we discussed. The other aspects, uh, to this would be, um, we see that customers are also looking at, uh, reducing the, let's say, the virtualization license cost, or improving the, uh, operational efficiency by moving into a unified, uh, platform so that, uh, a single, uh, team or a administrative group can manage both the, uh, workloads effectively, right? Um, so overall, we, um, we are looking forward that, uh, there could be more improvements and enhancements happening on this, uh, area of, uh, V M N containers convergence, uh, in the upcoming, uh, you know, uh, months.

So that's on the, uh, first trend, uh, that we discussed. So, following up, uh, the second, uh, uh, trend, what we are looking is running stateful, uh, workloads on, uh, container platform. So this use case is one other, uh, pattern that indicates, uh, you know, um, a cloud native platform, uh, becoming more reliable and, uh, preferred, uh, right?

So it is also, uh, connected with the previous, uh, one that we discussed, uh, the vm n container convergence where, uh, if you are going to scale the platform, so certainly, uh, you'll start, uh, consuming it for heterogeneous workloads. So in that way, uh, when your workload becomes versatile, so you'll end up in, uh, there's a need to host, uh, uh, stateful aware, the, the applications have a persistent data attached to it, um, so that, that becomes a need. So, uh, taking a step back, you know, um, by principle when, uh, containers started, so containers were short-lived and were built to execute, uh, uh, let's say a specific, uh, service or function, that by majority of the, you know, uh, enterprises started by hosting, uh, stateless, uh, applications are in other terms, uh, that does not need any persistent data attached to those, uh, applications, uh, enterprises, let's say.

Uh, it could be these applications could be, uh, crown jobs or periodic, uh, uh, functions or the stateless, uh, or web applications. So these were the primary, uh, targets or workloads that were hosted and with more, uh, consumption seen, uh, you know, we see that, uh, customers have started to evaluate, um, uh, the container of the Kubernetes platform to host, uh, stateful applications. So this means that, uh, you know, large enterprises have, you know, uh, already started, uh, hosting, uh, let's say tens or hundreds of, uh, database workloads on thousands of, you know, um, thousand nodes of, uh, Kubernetes clusters.

So that's the volume or even higher than that. So that's the volume that we are looking, the scale of running, uh, you know, straight, full workflows. We see that this would, you know, uh, even more accelerate.

So altogether, this, uh, paradigm could be a major shift in the, uh, cloud native journey as this will lead, uh, uh, to unlock the other, uh, you know, platform capabilities such as, uh, uh, you'll have to, uh, do the data protection for these, uh, workloads, and it has to be a cloud native way of, uh, protecting the data, uh, as the traditional ones may not apply here. And also, there should be recovery and resiliency policies attached to these workloads to make them know reliable and accessible and not highly available. So, for example, um, you know, uh, taking application, uh, application aware backup or application application, consistent backup, uh, floating, so, et cetera, those additional, uh, activities that happen, uh, for these container workload.

So some of the additional, uh, areas of focus, um, in this, uh, use case would be, uh, you know, identifying and defining the, uh, uh, use cases. What are the type of workloads to be hosted on this, uh, platform, and then getting them validated. And secondly, uh, designing and developing, uh, let it be the deployment of patterns and practices, uh, for running stateful workloads.

So, altogether, we, uh, the targeted, uh, benefits, uh, what we see are faster time to market by accelerating the application deployment, simplifying the, uh, application rollouts. And secondly, it would be lowering the infrastructure cost and effective, uh, utilization by assigning an a granular, uh, resource allocation models. And the third, uh, benefit could be, uh, improve, uh, developer and operator productivity, uh, with an automation first approach, um, for example, using, uh, operator of based control for the entire lifecycle management of these, uh, stateful workloads.

So that's on the, uh, uh, running stateful workloads and the data production that we discussed. So going forward, uh, the one other, uh, important or critical aspect is the, uh, cloud native, uh, security, now security for cloud native workloads. So, while this has been already there, uh, for some time now, security is the one major concern for, uh, container adoption, or, you know, in certain, uh, customers, we see that it has caused, uh, delay in taking applications to the production.

So, but there are growing, you know, uh, requirements in, um, you know, in streamlining the process with, uh, proper control mechanisms, uh, guardrails and, uh, governance in place. So considering, uh, these cloud native platform, uh, which will be existing, and, you know, it'll keep growing. So enterprises are, you know, started looking at, uh, investing, uh, uh, uh, creating a practice, uh, community, uh, to create an exercise, uh, you know, security best practices, uh, for better, uh, collaboration.

So one example, uh, you know, could be, uh, delivering, uh, uniform or a consistent, uh, container image. Uh, so it could be within the organization, uh, across all the business units, uh, so wherein, uh, container based images, uh, form the foundational layers of, uh, container and microservices, uh, application development. So, securing this base images is one crucial, uh, step.

Uh, let's say, uh, shifting left the security. Uh, so this helps, and starting the container security, uh, even before the, uh, you know, incorporating the application code into the container image. Um, but it gets extended throughout the life cycle of the application, including build, run, and deployed in the production environment.

So, um, uh, so basically the, uh, hardening this, uh, these basic majors, making them compliant as per, uh, you know, the, um, compliance standards like, um, uh, center of, in know intranet, uh, security, which is CIS benchmark or Department of Defense, which is d o d standard. So likewise, these standards, hearing the, uh, image compliance against these, uh, security standard. So, uh, streamlining the process of delivering, uh, these, uh, you know, hard end or, uh, the remediated images.

So we see this, uh, this is one prominent, uh, use case that's, uh, you know, um, that multiple customers are exploring. So, as a best practice, uh, you know, uh, enterprises also, uh, implement, uh, a DevOps or an integrated DevOps approach, um, in order to, you know, trigger and maintain this hardening the remediation pipelines, uh, process. So the, this is, uh, one example, the image hardening what we discussed.

So the other key focus areas in this space, uh, what we see is, uh, you know, securing the software supply chain, uh, starting from the, uh, code development, building, IT testing, and, uh, you know, even in the production deployment. And, uh, uh, the second, uh, would be defining and maintaining, uh, policies as a code. Uh, this is where we see, uh, open policy agent opa or, uh, berno, which is, uh, prominently used and for, uh, you know, port to port or container to container security microservices, uh, authentication and authorization.

So wherein, uh, the service mesh, uh, uh, providing, uh, uniform security compliance and policy poster. So these are the additional areas we see that, uh, you know, it's, uh, kind of picking up. So with that, um, next we would like to discuss on the cloud native, uh, the platform resiliency itself.

So, um, so far we discussed about the trends of, uh, you know, moving more, uh, workloads to the, uh, platform, uh, such as onboarding virtual machines to container running stateful workloads, and the proper security and governance in place. Um, so that right becomes important, uh, for the platform to be more, uh, agile and resilient, uh, to withstand, uh, failures. Uh, so for traditional workloads, uh, resiliency and disaster recovery, business continuity has all been well architected for decades now, which, or no many organizations are practicing it, but due to the nature of containers and microservices applying the, you know, the traditional approach may not, you know, suffice the business needs.

Uh, hence, there is a need to design, uh, resiliency and disaster recovery models for, you know, container, uh, specific, uh, workloads which we are observing. So for containers, uh, no, we see that, uh, resiliency is acquired at, uh, several layers. Uh, you know, it starts from the infrastructure where it can be on the on-prem, um, and of course, both, uh, physical or virtualized environment.

And if it is a cloud, uh, deployment, then the availability zone. So this is the infrastructure layer and followed by the cluster, uh, level that includes the, uh, master or worker or redundant, uh, you know, um, notes in order to, with, uh, self-healing, uh, capabilities. And of course, the final layer would be the containerized workloads itself, which is hosted on the platform.

So resiliency needs to be addressed at, uh, various levels. So in case of disasters or, you know, failures, the cloud native DR solution must, uh, suffice the, uh, critical business needs to quickly recover or restore the application, uh, from the secondary site. So this could in involve, uh, the modern, uh, software delivery approach like we discussed.

Um, you know, let it be infrastructure as a code or to rebuild the entire platform, or DevOps and GitHubs or approach to restore or quickly restore the applications, uh, at the DR site. So all this has to be well orchestrated within the, you know, the target limits of, uh, RTO and RPOs with, uh, proper, uh, automation workflows. So we see that traditional, uh, know the disaster recovery, uh, solution providers.

Um, now they have come up with, uh, started focusing on extending their capabilities to support these, uh, you know, use cases and features, um, that we discussed, right? Um, so with that, so these are the, you know, some of the common trends, uh, that I wanted to discuss. So, along with these trends, uh, you know, we have discussed, uh, uh, there could be one other, uh, initiative or, uh, you know, a trend that's, uh, being more explored is, uh, sustainability, right?

So whether, uh, containers can contribute to sustainability or how can they contribute, uh, to sustainability. So I just wanted to touch upon a few points around that. So, considering, uh, the characteristics of container, um, you know, uh, containers and assumably, the required, uh, uh, smaller, uh, computer sources, uh, footprint, um, to execute the, uh, application or service.

So, uh, innovate becomes, uh, you can efficiently utilize your underlying infrastructure. The hardware density increases by this, uh, the other, uh, uh, factors that containers are, uh, quickly scalable, so you can bring them down to zero and not in use, for example, um, in a non-production environment, you can scale it down to zero, uh, during the non-business hours so that, uh, it could contribute in saving some power and, uh, energy of those hard hardware. And, um, one other, uh, uh, contribution could be, uh, some of the open source projects like, uh, Kepler or Cube, uh, cube Green, um, where it is trying to estimate the, uh, carbon emission footprint, uh, per container level or pod level.

So, which, uh, no, it estimates. So based on that, you can have a governance over the, uh, carbon footprint. So, uh, this is also, you know, uh, has become, uh, one other element, uh, to consider cloud native, uh, you know, the container platform, the sustainability of fact, what we discussed.

So with that, uh, I think we have come to the, uh, end of this, uh, session. Uh, so to summarize, we discussed about, uh, the container and Kubernetes adoption trends, uh, which is increasing and becoming more, uh, prominent. And also we discussed about some of the, uh, you know, top use cases, what we are, uh, observing in the market, that in includes the, uh, VM and, uh, container convergence, uh, running stateful workloads on containers and cloud native, uh, security, and, uh, the platform resiliency itself, and of course, the sustainability initiative, what we discussed.

So with that, I hope the, uh, session was, uh, informative and useful. Uh, thank you for watching, and I look forward to seeing you in another informative session. Thank you.

Bye for now.