The emergence of cloud computing has opened up newer ways of sourcing IT services. It is the convenience of self-service, provisioning speed, virtually unlimited capacity and pay-per-use that made cloud computing popular. While large enterprises were slow in adopting public cloud with inhibitions around Information Security, Regulatory Compliance etc., startups and small/ medium companies took the bold steps forward and got extremely successful challenging many business models established by large enterprises, and proved that it all works.

Many enterprises didn’t want to own and manage datacenter facilities due to the huge CapEx required. In addition, building people resources to manage has also been seen as a challenge. These enterprises instead leverage the services of “CoLo” (CoLocated Datacenters) and “TOS” (Total Outsourcing) providers to host and manage their IT. The CoLo and TOS providers are also evolving and are going beyond “rack-space selling” to build and deliver “Managed Private Cloud” environments, bringing the benefits of public cloud environments into their service mix.

As obvious as it is, most enterprises want to leverage the best of all worlds – Private Clouds, CoLo, Managed Private Clouds and Public Clouds. This model of consuming services across different environments, breaking the boundaries of enterprise datacenter facilities, is typically called by many providers as “Borderless IT”, “BoundaryLess Datacenters”, “Pervasive Computing” etc. However, establishing, managing and monitoring such an expansive virtual boundary peeking into different cloud and IT service providers is easier said than done.

There are multiple factors that IT leaders need to take into account before building their “Pervasive IT” strategy:

Segmentation of application landscape: A careful assessment of the application landscape would be the first step in planning a Pervasive IT strategy. After gathering inventory, performance and dependency metrics of the applications landscape, it needs to be segmented into apps that needs to be retired, refactored or rebuilt to survive in a virtualized and distributed environment like the Cloud. It also needs to be segmented based on its Availability, Security and Performance requirements. A good cloud assessment toolkit can help identify cloud-ready apps.

Core vs ancillary infra services: Enterprises are actively looking at leveraging public cloud for ancillary infra services such as Disaster Recovery, Backup, Archival, Dev/Test etc. In a virtual environment, implementing disaster recovery in public cloud can significantly bring the costs down. As the storage costs keep going down, backup and archival also becomes reliable and cost-effective. In a cloud, Dev/ Test environments can be raised and shutdown in minutes, improving agility and rollback of environments convenient to Developers and Testers working on various branches of the code with different operating conditions.

Workload mobility: Based on the changing criticality of workloads, organizations may want to shuffle the placements of their workloads quite often. Critical workloads which needs to be on low latency networks may need to be placed on-premise. The same workload may lose its criticality after sometime and may be blocking expensive resources on-premise. Pervasive IT needs to have the capability to move workloads across cloud environments seamlessly, carrying along with them the security controls and policies.

Service Broker: In the Pervasive IT environment, where IT Services are sourced from multiple suppliers, IT leaders need to plan for a central brokering engine which can serve as a marketplace with catalog items from different suppliers, comparison of suppliers providing similar services, consolidated cost projects etc.

Security: When IT was limited within an enterprise datacenter, hardened security was implemented in the “perimeter” with all kinds of controls such as intrusion detection, prevention, firewall, anti-virus/ anti-spam and so on. As perimeters are broken in a Pervasive IT fabric, the security moves up to individual workloads. Each application is expected to protect its own availability and data security. However, the infrastructure needs to provide the foundational elements such as identity, authentication and authorization, roles and access etc. across different cloud environments.

Financial impact: It may not be the cost alone which drives public cloud adoption. In fact, there are studies that indicate on-premise datacenter services being cheaper than a comparable public cloud service. However, public cloud makes financial sense if the utilization is sporadic and not continuous. Below is a report from ISG:

Govindaraj Rangan- Head, Datacenter Innovation Office, Datacenter Practice, Wipro, Ltd

Govindaraj Rangan (Govind) is the Head of Datacenter Innovation Office, part of Datacenter Practice at Wipro. He has 19 years of industry experience across the breadth of the technology spectrum - Application Development to IT Operations, UX Design to IT Security Controls, Presales to Implementation, Converged Systems to Internet of Things, and Strategy to Hands-on.

Prior to Wipro, he spent over 10 years at Microsoft as Technology Strategist, working with some of the large Enterprise customers in India. He has also worked in the CIO/CTO organizations of Texas Instruments, Automatic Data Processing, D. E. Shaw & Co. and PCL Mindware.

He has an M.B.A. from ICFAI University specializing in Finance, M.S. in Software Systems from BITS Pilani and B.E. (EEE) from Madras University. Professionally, he is MCSE, CISSP, PMP, ITIL Foundation certified.