IT Operations Management (ITOM)
cancel

Inside story: How one customer transformed IT by building a cloud from the ground up

Inside story: How one customer transformed IT by building a cloud from the ground up

NimishShelat

Editor’s note: This article is part of an ongoing series of guest posts about HP Software customers Automation and Cloud Management use cases.

Advances in technology have driven some radical changes in recent years. No industry may understand this better than the entertainment industry, where digital, on-demand content is quickly altering how everyone does business.

 

This is the inside story of how one HP Software customer in the entertainment industry radically rethought its operations and built a cloud infrastructure from the ground up.

 

A Transformative Journey

This customer’s story begins three years ago, as it faced a major shift in how it managed media content. The customer moved away from physical media in favor of an all-digital supply chain to its partners worldwide. In order to support this new model, their IT infrastructure also needed to transition.

 

Like a lot of organizations then and now, “IT as usual” just wasn’t going to cut it. Although they had some virtualized IT infrastructure, it was not standardized; in many cases, deploying new infrastructure took an average of 6–8 weeks. As a result, their teams spent nearly 100 percent of their time on reactive issues and faced numerous infrastructure availability issues.

 

To stay competitive, a lot had to change.

The plan was ambitious—to build a robust, scalable and elastic infrastructure that would power their new digital supply chain.

 

The multi-tenant cloud had to support the varied requirements of multiple lines of business, with a much-improved security profile and 24/7 availability. Implementing self-service capabilities were essential to enable faster delivery of applications for the business, and their goal was to reduce infrastructure provisioning time from eight weeks to within an hour. Their cloud, as they envisioned it, would benefit multiple groups, including their internal IT and customer infrastructure IT shared services, as well as their developers.

 

But making such a major shift required change on several levels, including personnel reorganization, new operating processes, a fresh approach to service management and even a new financial model.

 

An End to Technology Silos

Like most organizations, this customer operated within technology silos, but they quickly realized that supporting a cloud platform required more integration and a stronger focus on process, not just raw technology. They reorganized their teams to become a process-oriented organization, with integrated teams in four general areas:

  1. Plan — Business Management, Architecture
  2. Build — Projects, Engineering, Integration
  3. Transition/Transform — Release Management, Deployment Automation
  4. Run — Operations, Applications Support, Service Support

 

The integrated teams have made problem resolution more natural, while proactive activities have a consistent, structured flow.

 

Adopting new processes was itself a challenge. Before, building a server required several manual and disconnected processes as well as coordination with many different teams. If they were going to succeed with their cloud transformation, they knew it had to be a single integrated and automated flow. Suddenly things like standards and capacity management — disciplines they were not adept at — became critical for making cloud work.

 

Standardized infrastructure also played a big role. To provide maximum provisioning flexibility, they selected HP storage and HP servers. In addition, they implemented HP Cloud Service Automation (CSA), Operations Orchestration, Server Automation, and Database and Middleware Automation. Like most IT environments, they have multiple sets of tools in use, but all of the HP tools integrated nicely with third party tools for on-going individual configurations and patching.

 

Service Management of the Cloud

 

One of the key aspects of their cloud implementation is that they require the user to pick or associate the server with an application right upfront. This approach prevents a lot of undocumented disorganization in their CMDB. Now HP CSA passes all the information to the CMDB when the client orders a server, and ultimately that gets pushed into their service desk on the backend to ensure the CMDB stays up to date.

 

This plays into their service management architecture, where servers or components make up applications and in turn, applications make up business services. These services can then feed into tools for application monitoring.

 

Cloud Service Management.png

 

Fig. 1: Logical diagram of cloud service management

 

Move from Allocation financial models to Rate Card Services

Infrastructure and platform services are now both accessed via a web-based service catalog, which comes with a new financial billing model for Shared Services. This was an important aspect of their transformation. Like many traditional shared services organizations, the company previously used a completely allocated model. Although users were used to it, it was unpredictable and pushed all the work to the end of the year.

 

But with the rise of cloud and market rate services, they needed to match these new models for resource consumption.

Today, their customers are quickly becoming conditioned to Rate Card Services, which have the benefit of being a much more predictable pricing model. Although it’s definitely harder in some respects — it puts majority of the work up front in calculating the rates — they’ve found the change to be worth it and matches the new expectation from anyone that has used public cloud services.

 

Results

 

So, what did the customer achieve by making all these changes?

First and foremost they were shooting for 1-hour server builds, but actually got down to 15 minutes. They also now have availability in the range of 99.99 percent and were able to consolidate previous infrastructure from five racks of gear, on average, down to one with HP’s dense computing platforms.

Security has improved through more granular network segmentation made possible by automation, and across the board, agility has improved.

 

Finally and most importantly, the consolidation enabled dramatically lower OPEX costs.

 

Recommendations for cloud building

Is your organization is considering a wide-scale adoption of automation and cloud for your IT environment? Here are a few tips based on this customer’s experience so far:

 

  • Think strategically — Consider all of the potential beneficiaries of your new infrastructure, so that you can plan to accommodate their needs and get the most value out of the new environment. You definitely need to have that big vision and a roadmap for where you want to go.
  • Narrow your scope for faster wins — Initially, keep your scope pretty tight. It is easy for projects like this to have scope creep in the beginning. Cloud can get big and complicated quickly, so break it down with frequent, measurable progress to maintain momentum.
  • Reuse code — This is important to keep in mind when you are building processes.
  • Go agile — Waterfall doesn’t really work, so adopt some type of Agile methodology
  • Relentlessly pursue standardization — This is critical for driving scale and ROI, while customization is, in many cases, the enemy of cloud
  • Integrate security — In most cases, it’s easy to automate something. But to do it securely usually takes extra work; take the time to do it right

 

What’s next?

This customer is moving quickly. With IaaS deployed and working, the PaaS offerings came out of beta in the late spring and the customer is now at the beginning stages of implementing software-defined networking. Soon they hope to tackle storage-as-a-service, too.

As they finish up their initial cloud deployment, their next plan is to integrate public cloud services for hybrid scenarios, and deploy a next-generation on-premises Helion OpenStack environment to give developers an API-driven native cloud platform.

It’s an impressive transformation over just three years, and one that can serve as a model for how implement cloud IT successfully.

 

Learn more

Read more about HP Automation and Cloud Management (HP ACM) here.

  • infrastructure management
0 Kudos
About the Author

NimishShelat

Nimish Shelat is currently focused on Datacenter Automation and IT Process Automation solutions. Shelat strives to help customers, traditional IT and Cloud based IT, transform to Service Centric model. The scope of these solutions spans across server, network, database and middleware infrastructure. The solutions are optimized for tasks like provisioning, patching, compliance, remediation and processes like Self-healing Incidence Remediation and Rapid Service Fulfilment, Change Management and Disaster Recovery. Shelat has 23 years of experience in IT, 20 of these have been at HP spanning across networking, printing , storage and enterprise software businesses. Prior to his current role as a Manager of Product Marketing and Technical Marketing, Shelat has held positions as Software Sales Specialist, Product Manager, Business Strategist, Project Manager and Programmer Analyst. Shelat has a B.S in Computer Science. He has earned his MBA from University of California, Davis with a focus on Marketing and Finance.