As a provider of both application development management and infrastructure outsourcing, Denmark-based NNIT needed a better way to track, manage and govern more than 10,000 services across its global data centers.
Beginning in 2010, the journey to better overall services automation paved the way to far stronger cloud services delivery, too. NNIT uses HP Cloud Service Automation (CSA) to improve their deployment of IT applications and data, and to provide higher overall service delivery speed and efficiency.
To learn more about how services standardization leads to improved cloud automation, BriefingsDirect spoke with Jesper Bagh, IT Architect and cloud expert at NNIT, based in Copenhagen. The discussion, at the HP Discover conference in Barcelona, is moderated by me, Dana Gardner, Principal Analyst at Interarbor Solutions.
Here are some excerpts:
Gardner: Tell us about your company and what you do. Then, we’ll get into some of the services delivery problems and solutions that you've been tasked with resolving.
Bagh: NNIT is a service provider located in Denmark. We have offices around the world: China, Philippines, Czech Republic, and the United States. We’re 2,200 employees globally and we're a subsidiary of Novo Nordisk, the pharmaceutical company.
My responsibility is to ensure for the company that business goals can be delivered through functional requirements and in turning the functional requirements into projects that can be delivered by the organization.
We’re a wall-to-wall, full-service provider. So we provide both application development management and infrastructure outsourcing. Cloud is just one aspect that we’re delivering services on. We started off by doing service-portfolio management and cataloging of our services, trying to standardize the services that we have on the shelf ready for our customers.
That allowed us to then put offerings into a cloud, and to show the process benefits of standardizing of services, doing cloud well, and of focusing on the dedicated customers. We still have customers using our facility management who are not able to leverage cloud services because of compliance or regulatory demands.
We have roughly over 10,000 services in our data centers. We’re trying now to broaden the capabilities of cloud delivery to the rest of the infrastructure so that we get a more competitive edge. We’re able to deliver better quality, and the end users, at the end of the day, get their services faster.
We embarked on CSA together with HP back in 2010. Back then, CSA consisted of many different software applications. It wasn't really complete software back then. Now, it’s a full suite of software.
It has helped us to show to our internal groups—and our customers—that we have services in the cloud. For us it has been a tremendous journey to show that you can deliver these services fully automatically, and by running them well, we can gain great efficiency.
Gardner: How has this benefited your speed-to-value when it comes to new applications?
Bagh: The adoption of automation is an ongoing journey. I imagine other companies have also had the opportunity of adopting a new breed of software, and a new life in automation and orchestration. What we see is that the traditional operations divisions now suddenly get developers trying to comprehend what they mean, and trying to have them work together to deliver operations automatically.
Back in the good old days, developers were in one silo, and operations were in another silo. Now, we see a mix of resources—both in operations and in development. So the organizational change management derived from automation projects is key. We started up, when we did service cataloging and service portfolio management, by doing organizational change to see if this could fit into our vision.
Gardner: Now, a lot of people these days like to measure things. It’s a very data-driven era. Have you been able to develop any metrics of how your service automation and cloud-infrastructure developments have shown results, whether it’s productivity benefits or speeds and feeds? Have you measured this as a time-to-value or a time-to-delivery benefit? What have you come up with?
Bagh: As part of the cloud project, we did two things. We did infrastructure as a service (IaaS), but we also did a value add on IaaS. We were able to deliver qualified IaaS to the life science industry fully compliant. That alone, in the traditional infrastructure, would have taken us weeks or months to deliver servers because of all the process work involved. When we did the CSA and the GxP Cloud, we were able to deliver the same server within a matter of hours. So that’s a measurable efficiency that is highly recognized.
Gardner: For other organizations that are also grappling with these issues and trying to go over organization and silo boundaries for improvement in collaboration, do you have any words of advice? Now that you've been doing this for some time and at that key architect level, which I think is really important, what thoughts do you have that you could share with others; lessons learned perhaps?
Bagh: The lesson learned is that having senior management focus on the entire process is key. Having the organization recognized is a matter of change management. So communication is key. Standardization before automation is key.
You need to start out by doing your standardization of your services, doing the real architectural work, identifying which components you have and which components you don't have, and matching them up. It’s trying to do all the Lego blocks in order to build the house. That’s key. The parallel that I always use is there is nothing different for me as an architect than there is for an architect building a house.
Gardner: Looking to the future, are there other aspects of service delivery, perhaps ways in which you could gather insights into what's happening across your infrastructure and the results, that end users are seeing through the applications? Do you have any thoughts about where the next steps might be?
Bagh: The next step for us is to be more transparent to our customers. So the vision is now we can deliver services fully automatically. We can run them semi-automatically. We will still do funny stuff from time to time that you need to keep your eyes on. But in order for us to show the value, we need to report on it.
The next step for us is to be more proactive than reactive in our monitoring and reporting capabilities, because we want to be more transparent to our customers. We have a policy called Open and Honest Value-Adding. From that, we want to show our customers that if we can deliver a service fully automatically and standardized, they know what they get because they see it in a catalog. Then, we should be able to report on it live for the users.