A common question from my clients is, “How do I transform my existing data center services into services fit for the 21st century?”
In order to answer this I have to understand what 21st century services “look” like in an ideal world. I believe 21st century like services should enable the following:
- Charging for capacity used instead of capacity available
- Transparency of billing and metering amounts
- Easy enabling of new services
- Fast and easy provisioning of new environments and de-provisioning of environments no longer required
- Proactive monitoring to be able to react when a problem is about to occur rather than after the event
- Predictive analytics to enable improved future planning requirements
- Self-service capability
- Standardization across the enterprise
The hard part, for the majority of my clients, is that they cannot just build a new data center capable of 21st century–like services; instead they need to transform their existing data center (including services) while still maintaining required service levels on their production systems.
So how can they make the transition? There is no right or wrong answer, but Smarter Computing (and especially the cloud capability within Smarter Computing) can definitely be an enabler in this particular situation. You can definitely achieve step-change improvements when using cloud solutions.
In order to attain these step-change improvements and ultimately achieve a data center with 21st century–like data center services, you need to consider and address the following factors when deciding on the approach to follow:
- The transformation is a journey and cannot be achieved using a big-bang approach.
- No one size fits all; instead a range of cloud (and non-cloud) based solutions and services are required to match the varying requirements and demands of client workloads.
- Consolidate and map workloads to the desired cloud delivery model (see an example workload to cloud delivery model diagram below) keeping in mind that not all workloads will map directly to a cloud delivery model; that is, it is not unusual to have exceptions.
- Plan to automate, virtualize and standardize as much as possible without compromising system functionality.
- Integrate existing monitoring tools to provide dashboard-type view and/or investigate improved monitoring tools to provide more proactive capabilities.
- Utilize already captured audit, logging and monitoring data and integrate with trend analysis and predictive modeling tools to enable predictive reporting rather than “after the event” reporting.
- Consider if transparency of billing and metering should be used as an enabler to implement business recharge models.
- Decide on limits of flexibility provided for self-service, especially around provisioning and de-provisioning of environments.
- Security, security, security: ensure that the balance is correct between enabling flexibility around self-service versus authentication, authorization and integrity of all systems, and ensure that security policies exist (and are adhered to) for both internal and external systems.
- Remember that this is not just a transformation journey for technology and services but also for people, so ensure that processes and procedures are revamped and that people are trained in how the new model works.
Due to evolvement of cloud solutions and the differing cloud delivery models, it is critical to have strong cloud service integration capabilities, governance and management.
The end result, hopefully, will be successful delivery of data center services fit for the 21st century.
Siobhan Nicholson currently works as a Client Technical Advisor for Consumer Product accounts within the UK. This includes building an understanding of client environments, identifying technical challenges, and leveraging IBM’s technical resources for the benefit of clients. You can find Siobhan on Twitter @SUN_gator and LinkedIn.
To effectively compete in today’s changing world, it is essential that companies leverage innovative technology to differentiate from competitors. Learn how you can do that and more in the Smarter Computing Analyst Paper from Hurwitz and Associates.