Intel Speeds Delivery with a Hybrid-Cloud Strategy

Keeping up with the rapid changes in technology is increasingly difficult, especially for organizations with existing IT infrastructure. In my 20 years of enterprise solution design experience, I’ve seen how difficult it can be to make a digital transformation to the cloud. But with a well-informed strategy, cloud services can support business acceleration. At Intel, we grappled with the same challenges most organizations face: optimizing our data centers to accelerate application delivery and better meet customer needs. We developed the Intel® Cloud Application Platform to meet these needs and easily create, deploy, and manage applications. With over 2,000 applications and micro-services on our internal cloud environment, our cloud strategy has allowed us to move more quickly while reducing costs. Through this process we learned the importance of developing a cloud strategy based on our own specific needs, both business and technical. With a multi-cloud approach, we have achieved significant operational improvements and helped accelerate business. You can learn more below and in our detailed paper Optimal Workload Placement for Public, Hybrid, and Private Clouds.

I have found that there is no one-size-fits-all cloud strategy. At Intel we identified the following three-step approach that helped us determine which cloud services we needed—public, private, or hybrid:

  1. We first identified our business considerations, such as financial strategy, end-user location, regulatory compliance, and service-level agreements.
  2. Next, we identified the technical requirements for each of our workloads, such as performance, security, and integration needs, as well as data volumes.
  3. Finally, we reviewed our ecosystem options, such as the maturity of solutions available and the level of effort needed to migrate existing solutions.
Optimal Workload Placement Infographic
(Click image to view full-size)

Intel’s cloud strategy includes deploying public software-as-a-service solutions, a private platform-as-a-service (PaaS), and database-as-a-service (DBaaS) capabilities. The applications most central to our business are less appropriate for public cloud, including public PaaS. For some of these applications, the performance criteria require solutions that are closer to the end-user's internal systems. But integrating existing solutions with external environments often extends the application development and delivery by weeks or months. For these reasons, we decided to deploy our own set of developer services through PaaS and DBaaS to enable applications and line of business, hosted within our private cloud infrastructure.

As we identified our business considerations, one requirement was that the solution enables business agility. The primary business benefits of PaaS were decreased delivery time for custom applications through shared services, increased pace of development due to a lighter administrative burden on developers, improved integration with legacy solutions, lower infrastructure costs due to location independence and shared services, and improved security compliance through platform-enabled efficiencies.

With our PaaS cloud strategy, we increased the efficiency of resources for IT. Compute and density was increased by switching from a single-tenant to multi-tenant hosting, which led to lower support costs. Fewer personnel are required to support the application and database infrastructure.

Our Cloud Strategy Delivered Results

By implementing a combination of cloud solutions that targeted our workload and business needs, we achieved the following:

  • Data center improvements. We improved overall storage practices, scale-out storage, storage refresh, and reduced data duplication while achieving a smaller overall footprint.
  • Design environment improvements. We achieved more efficient design compute and storage, improved high-performance computing, increased design throughput, and optimized servers to meet increasing compute demand.
  • Office and enterprise compute and storage improvements. We increased the number of virtual machines from 11 in 2009 to 26 in 2015 to meet the increasing demand for compute and storage from the office and enterprise customers.
Optimal Workload Placement for Public, Hybrid, and Private Clouds whitepaper
(Click to download white paper)

To learn more about our approach and how it can help any business develop a cloud strategy, read the white paper Optimal Workload Placement for Public, Hybrid, and Private Clouds or view our infographic.