HP's Robin Purohit Unpacks Business Service Management 9 as Way to Address Complexity in Hybrid Data

Updated: June 12, 2010

Purohit: Our customers are dealing with some of the most significant combination of changes in IT technologies and paradigms that I had ever seen. There's a whole new way of developing applications like Agile development, the real acceleration of virtualization from your desktop and test environments to production workloads, and all the evaluations of where the cloud and software as a service (SaaS) fits and how that supports the enterprise applications.

All these things and more are colliding at once. What customers are saying is, "How do we take advantage of these new technology shifts and new ways of dealing with technology to get dramatic impact in cost, but not increase the risk? How can I do things faster and cheaper, but do things right?"

What they're looking for is a way to somehow simplify and automate the use of all these technologies and their processes using management software, so they get the most they can of these new paradigm shifts, while they keep up what the business wants them to do.

Compared to a year ago, the active interest of our clients in cloud computing has just exploded. Last year was a curiosity for many senior IT executives. It was something on the horizon, but this year there's really an active evaluation.

Most customers are looking initially at something a little safer, meaning a private cloud approach, where there is either new stack of infrastructure and applications are run for them by somebody else on their site or at some off-site operation. That seems to be the predominant paradigm.

Piece of a puzzle

The challenge is that that set of cloud services, that private service, is really just a piece of a puzzle to run their business operation. It's usually slice of infrastructure or certain class of application that's part of the larger critical business service for their company.

What they have to do is to figure out how to take advantage of that, target the right workload where it's okay to take that risk, select the right partner, but then make sure that all of the instrumentation of making sure they are getting what they wanted out of it is actually integrated with the rest of their operation. Otherwise, it's just another thing to manage.

The challenge for IT is how they enable that level of innovation at the right pace, but make sure that it's all very well governed -- simple things like getting what we pay for in terms of performance, capability, and the capacity.

That whole notion of cloud governance is one of the most critical things to get right near term, so that the IT guys can keep up with the business guys.

We're sourcing some sort of elastic-like services. And, by the way, is that environment secure, so we're not putting the business at risk? Then, if I want to change to a different cloud provider or another private provider, how do I do that in a fairly nimble way, without having to re-architect everything that I have done.

That whole notion of cloud governance is one of the most critical things to get right near-term, so that the IT guys can keep up with the business guys, but all the risk is there, it's still going be on the line.

There are two really big cost drains in IT. One is that when they roll out your applications, the majority of the time, things don't work well when they initially roll something out. If you think of Agile and the pace at which now new application innovations are bring rolled out, it really means that you have to get things right the first time. The first thing is to tackle that problem, as you go to these hybrid models.

Chaotic environment

The second thing is that most companies still are trying to get a handle on a right way of simplifying and automating their operation in a very chaotic environment. A typical data center is dealing with 900 changes a month. They might get a million incidents over a couple months and each one of those incidents could cost up to $80 plus labor. So, you can just imagine how chaotic and expensive it is just to run day-by-day.

It's really critical for both, the hand-off of the applications to operation, as well as this running the daily operations. This is incredibly automated and simplified and all focused on the impact of all these things that the business wanted in the first place.

If we do that right and then make extensions into all of these same core processes to accommodate a SaaS model, a private cloud model, or even ultimately a public cloud model, without having to change all of that, you are going to be able to bridge from today to the future. You'll be getting all that benefit and actually keep reducing your cost, because you want to keep doing this innovation in a sustainable way.

Arrival of HP BSM 9.0

This has been a great release, and we're incredibly proud of it. BSM 9 is our solution for end-to-end monitoring of services in the data center. It's been a great business for us, and we have a break-through release that we revealed to our customers this week.

It's anchored on what we call the runtime service model. A service model is basically a real-time map of everything from the business transactions of the businesses running to all of the software that makes up that composite applications for the service, and all of the infrastructure -- whether it be physical or virtual, on-premise or off-premise -- that supports all of that application.

All of that together -- knowing how it's connected, what the health of it is, what's changing in it so, you can actually make sure it's all running exactly the way the business expects -- is really critical.

If you can imagine what we've talked about with virtualization and the rate of change there, people optimizing virtual workloads, new application coming on as being fired in the data center with Agile and maybe some outsourced environments and private/public clouds, that service model better be real time and up to date all the time.

That's the real break through. Before we had a service model that was really linked to the configuration that we thought was running. Now that they have everything up to date in real time with all of this increased velocity, it's really critical.

So, we've rolled that out, and it's now the backbone for all of our end-to-end monitoring. The other thing I'd stress is that once you have that, especially, in this very fast-paced environment, you can really increase the levels of automation.

What you've seen before

When you detect an event, making sure you know exactly what was going on at the time of the event, you can help people diagnose it and probably help solve it, because most of the times these things are based on what you've seen before.

We've taken all of our world-class automation technology, wrapped right into this end-to-end monitoring solution to automate everything possible. We think this can drive dramatic reduction in the cost of operations.

The last thing, I'd emphasize is that there are a lot of people involved in solving these problems, and running these operations. What's important is that all of them have a very personalized UI that looks and feels like a modern application, but os all based on one version of the truth of what's going on. We made major improvements, just overhauling the way all of this is presented in a very rich Web 2.0 way, but also in a way that's targeted to the needs of every single user in operations.

What we're trying to do in HP is not just worry about the data center. We're trying to help customers really adapt and morph these applications into the new world. Most customers are shifting their IT focus on innovating around the application. This means that more of their people who they have internally are creating new IT in the form of a new application for a sales person or new customer-facing portal. That's going to drive more revenue.