To be a strategic partner with the business, IT needs to focus their cloud strategy on the applications that deliver the business value; not the infrastructure. Easier said than done. Or is it? Lets talk to the experts and find out…
Most IT organizations are executing the 2nd or 3rd iteration of their cloud strategy. Some Global 2000 IT organizations that support multiple business units may even be dealing with multiple cloud strategies.
- How can you, as an IT leader, set and execute a cloud strategy today that can change and evolve in the future?
- How does an “App 1st” approach help keep your organization focused on business value and not drive decisions based on the infrastructure?
Executive Interview Series
This is the first in our Cloud Strategy Brief blog series, including interviews with IT executives who are responsible for setting and executing cloud strategy. By learning from the experiences of others, my hope is that you can experience “the gain without the pain,” and skip some of the trial and error steps along the way.
Gaurav Manglik – CEO and cofounder of CliQr Technologies
Q: What was your vision when starting CliQr Technologies?
Our vision was always to create a very app-centric cloud management platform. Businesses have always realized that their applications drive value. The challenge has been that infrastructure has been required to deliver applications to the business.
In the past, building and running the infrastructure was a major challenge. Infrastructure was a hardware investment – you had to stand up infrastructure first, and then deploy applications on top of that to drive return on that investment.
The cloud changes this equation. With the cloud, you have options that aren’t tied to an infrastructure investment. It’s now possible to configure infrastructure in an agile way for the very first time in the history of IT. But applications are still being deployed in ways that are coupled to the underlying infrastructure.
I saw the industry quickly moving beyond the infrastructure challenge, with cloud becoming an inevitable part of the IT landscape. We set out to create a solution that would decouple the application from the cloud. And, saw the application and application decisions becoming the logical construct that drives the process, not the infrastructure. And that insight led Tenry and I to start CliQr.
Q: Is that why CliQr emphasizes applications over infrastructure?
If the business value is delivered by the applications, you want to make application decisions first, and determine the right ways to deploy those applications – what kind of SLAs, what kind of quality of service, what kind of performance and latency requirements, what usage or placement policies need to be enforced – and then make placement decisions.
With cloud, you can control infrastructure through software “knobs”, and deploy and manage applications on premises or in a public cloud. So now, it’s actually possible with knob-controlled infrastructure to make the application the main part of the decision equation, rather than the other way around.
IT can help find the best deployment options for applications. But those decisions should largely be hidden from developers and users.
Q: As founder of a startup, you need to “skate to the puck” so to speak. Are you confident in your initial investment thesis?
Never have I had more conviction around our initial thesis. And the thesis went back to two key observations. First, to get applications up and running and gain agility benefits of the cloud, you have to provide automation. Automation turns the cloud infrastructure “knobs.” The second observation is that the cloud “knobs” are all different. Private and public cloud APIs are all different. AWS is different than OpenStack. Azure and Google, are different than VMware.
When I started CliQr, it was still an open question whether the industry convergence was on a single interoperability standard. That didn’t happen. Each cloud service provider has a vested interest in making their cloud different. And there have always been automation solutions. But focusing on the application and deploying automation that is cloud agnostic is the core challenge.
That is the problem we set out to solve. We assumed that clouds would all be different; that our customers would want to leverage different clouds for different uses, and IT would want to be the primary IT service provider for their business, finding the best fit for workloads, either in their datacenter or in the public cloud.
Our approach gives customers the ability to focus on the application, and leverage the power of the cloud without the risk of lock-in. So yes, we have skated to the puck.
Q: Are customers buying it?
Yes. We have definitely seen our application-centric approach validated in the marketplace. We have a great set of customers that tend to be large enterprise IT organizations. I’m impressed by the wide range of use cases they have deployed.
Several things are driving adoption. One is diversity in application types. Another is diversity in cloud environments. Another is the need to work across environments during different stages of the application lifecycle.
So that diversity in applications, that diversity in what you’re planning to do with those applications, and the diversity in cloud environments is driving adoption of application centric cloud management.
Q: How does CliQr help those responsible for overall cloud strategy?
IT decision makers are faced with a wide range of private and public cloud options that are all technically feasible. And the number of technologies is still rapidly expanding. But we are still early in the cloud adoption cycle.
So a cloud strategy that makes sense today may not in 18 or 24 months. Requirements or constraints may change. New technology may emerge. I think the only thing an IT executive can count on is that their cloud strategies will evolve over time.
So “future proofing” a cloud strategy is really important. There shouldn’t be a big cost to deploy an application to the cloud. There shouldn’t be a penalty if you change your mind and bring it back, or move to another cloud. But that is what many IT decision makers are faced with if they don’t have the right solutions in place.
For example, we have a fortune 500 customer that deployed a complex business critical application to AWS. Plans changed, and they needed to move that application stack to the Google cloud. Plans changed again and they moved that application stack to Azure. Those changes weren’t part of their original cloud strategy. But that type of change is the reality that IT decision makers need to prepare for.
You don’t want to have to present to the board of directors a spending request that is needed to undo last year’s spending request. Our application-centric cloud management solution helps give IT decision makers flexibility with their current cloud strategy, and gives them peace of mind that they won’t get locked into a specific approach and pay a change penalty.
Q: Are there other factors to consider when considering cloud strategy?
I think a “One cloud” approach isn’t going to fit reality. A strategy where a datacenter is seamlessly extended to a single public cloud offering is an interesting idea. But just doesn’t fit where the industry is headed. We may see consolidation in the number of public cloud providers. But there will always be new, better ideas that offer options at the margin that will make room for new providers. That is what drives our industry.
I understand why many industry incumbents have a hybrid cloud offering that fits their business interests. VMware wants to connect vSphere customers to vCloud Air. Microsoft Systems Center to Azure. Red Hat wants to connect Openstack to one cloud AWS. With that approach, you still are fitting applications to a specific infrastructure or cloud service. And, in many cases, they are adding public cloud connection, and integration to existing management stacks that were originally written for on premises management. That creates complexity.
And that approach doesn’t fit reality. We have customers who have six, going on eight different private and public clouds to manage. That’s 20 plus management zones. Each might be configured differently, with different compute, storage, networking, different infrastructure layers, different APIs. Each offers a unique value as an execution engine for different applications needs. But if each has it’s own team and management stack, you loose flexibility, control and increase costs.
Q: So what is the alternative?
A federated cloud strategy is a much better approach. By choice or not, many IT executives have quickly moved beyond the stage where they have one cloud as an extension of their datacenter.
For example, you might have a mission-critical, security-sensitive workload that can only run on a private cloud infrastructure because it meets your compliance and security requirements. However, there might be a different set of applications that run better on a public cloud. Or, you might use cloud for that sensitive workload’s development, but run production in your datacenter in a VMware environment.
That diversity in applications, in what you’re planning to do with those applications, requires a federated approach to managing applications in different environments.
Q: What type of projects offer a good place to start?
We have customers that don’t yet have a hybrid cloud strategy. They start by migrating a few key applications to a single cloud, they offer self-service on demand provisioning of key application stacks in their datacenter, or they integrate that automated app stack provisioning with their development tools for automated dev and test workflow. But, they share our vision that a multi-cloud strategy is the likely outcome. By starting with an application centric platform that works seamlessly across their datacenter and the public cloud, they have flexibility to start small and grow as their needs change.
We have other fortune 500 customers who are already dealing with a broad portfolio of multiple private and public cloud environments. They are looking for flexibility to move applications from one environment to another, without rewriting the application or the deployment scripts that are specific to each environment. They are automating dev and test across environments. They also are delivering IT as a Service, and brokering cloud deployments as a centralized IT service.
Q: Any words of advice for IT executives working through their cloud strategy?
Sure. Keep your options open. We are in a period of rapid evolution. Don’t bet on a “one cloud” future. Don’t get locked into a cloud strategy you know will evolve and change. And, look for a single management platform that works in your datacenter and across multiple clouds – one that’s designed to abstract the application from the cloud environment, and that supports a wide range of uses. Choose CliQr CloudCenter.
Subscribe to our blog to keep up with future Cloud Strategy Brief blog posts.