Delivery and Adoption of Cloud Computing

Michael Stoeckert, CTO, ProAssurance [NYSE:PRA]

343

595

118

Michael Stoeckert, CTO, ProAssurance [NYSE:PRA]

What is “Cloud?"

Cloud computing delivery models provide users with on-demand access to a flexible, wide-ranging pool of technology assets composed of services, applications, servers, networks, and storage facilities.

Due to the habit of referring to data and applications as being “in the cloud,” it can be easily forgotten that cloud is housed in physical facilities. Cloud computing services are powered by highly virtualized processing and storage systems in data centers, which are available to remote users via the web.

Clouds consist of groups of one or more physical servers sitting behind an access point that distributes requests for entry. Proprietary hosted systems can be built in-house on top of existing hardware; or access to external clouds can be rented on-demand from third-party vendors such as major online retail companies. The cloud’s virtualization feature allows users to run multiple applications and operating systems independently on a single server. Additionally, administrators can quickly move workloads from one virtual workspace to another, easily prioritizing business needs while maximizing server resources.

How Does Cloud Computing Work?

Clouds provide request-driven, dynamic allocation of computing resources for a mix of workloads on a massively scalable virtual infrastructure. The user of an application can request resources from the cloud and obtain them quickly; the cloud computing model reduces the need for capacity planning at an application level. By contrast, in a non-cloud/non-virtualized environment, it could take much longer for administrators to restart a server or change hardware or software conﬁgurations.

In many industries, a public cloud is the most recognized form of cloud, but private clouds dominate regulated industries

The Evolution of Cloud Computing

The foundation of cloud computing was laid in the 1980s. From the grid computing mainframes of that and previous decades, utility computing then offered clusters as virtual platforms with a metered business model by the 1990s. More recently, software as a service (SaaS) has raised the level of virtualization to focus on everything from subscriptions to applications. From those origins, cloud computing allows today’s users to gain access to their applications from anywhere, at any time, through a wide variety of portable devices.

Different Clouds for Different Needs

In many industries, a public cloud is the most recognized form of cloud, but private clouds dominate regulated industries. It is important to make the distinction. Rather than having dedicated equipment for each client (servers, networking, surge protection, storage, bandwidth, etc.), public clouds provide a shared pool of resources. By sharing equipment, public cloud clients can obtain the greatest possible cost efficiencies, as well as excellent scalability and elasticity.

However, because a public cloud does not have the same regulatory scrutiny, security can be a concern. Shared internet connections and shared servers typically provide access—this is less than ideal for hosting sensitive applications and data.

For increased security, many businesses turn to a private cloud solution. In a private cloud, all resources in the operating environment are dedicated to a single client. This ensures the client’s data never shares virtual space with another company. With private interconnects or multiprotocol label switching (MPLS) circuits and virtual private networks replacing internet connections, clients can also be granted access to security and management controls as needed.

In many cases, companies are starting to explore a hybrid cloud approach, connecting data centers and public and private clouds in any combination. Some firms use a public cloud for non-sensitive data or testing and rely on a private cloud for critical data and applications. The individual clouds stay independent, but are bound together to facilitate the portability of data and applications.

With a hybrid cloud, an organization manages some resources in-house while other functions are managed externally. For example, an organization might choose to migrate their email function to the cloud, while performing other tasks requiring a high level of integration (core systems like policy, claims, loans, deposits, logistics, and order processing) within their own IT infrastructure. Also, an organization can use the private cloud for backups of data and disaster recovery instead of maintaining an IT infrastructure at a secondary location.

The biggest challenge of moving to a private cloud is if your systems are not stand-alone but highly integrated. This is due to latency that will occur between two integrated systems when one is on premise and the other is in the private cloud. To reduce latency and meet user performance expectations, you will need high-speed interconnectivity which will offset much of the savings. If you have a long-term plan to move the entire integrated system to the private cloud, remember to not count on many cost savings until you can remove the high-speed interconnect.

For many firms, the hybrid approach can be the best of both worlds. It allows a business to take advantage of the scalability and price of a public cloud and dedicated capabilities of highly-secured environment in a private cloud.

Conclusion

Today’s realities make cloud computing a logical ﬁt. It meets the needs of businesses and their clients, who demand ﬂexibility, efficiency, and support for the completion of large workloads at high speed.

Cloud computing can represent the cornerstone of a powerful business strategy—combining reduced costs and increased regulatory responsiveness, while maintaining a high level of security. With cloud computing, companies can redeploy valuable resources to undertake more business development initiatives.

However, companies need assurance that cloud provider(s) have the depth of knowledge and experience required for firms that operate decisively in today’s markets and ecosystems.

Many companies will enter into cloud computing with a goal of saving money. They will quickly see that the cloud is an entry point to increased efficiency through the evolution of technology architectures, management tools, and operational processes.