|By Ajay Budhraja||
|September 14, 2012 08:00 AM EDT||
There has been a push for Shared services adoption across organizations for a while since such services reduce duplication and can result in cost savings. However there have been challenges to proposing and implementing shared services. Shared services initiatives challenges have been related to areas such as resistance to change, start up costs and management of such services. Cloud is enhancing the pace to shared services migration and it naturally facilitates the use of shared services since a cloud service can be more easily leveraged by multiple consumers.
Cloud is the true manifestation of a service delivery mechanism and has significantly sped up the transition to consolidation and shared services. Cloud can be termed as the next generation of shared services since it adds the dynamic computing, elasticity, self-service, measured aspects in addition to other aspects for rapid provisioning and on demand access. Cloud solutions may offer lower lifecycle costs based on usage and the monitoring aspects can lay out a holistic view of usage, cost assessments and chargeback information. All this information can enhance the ability of the organization to plan and react to changes based on performance and capacity metrics.
In several of my keynote presentations, I have emphasized the value of having a sound service based foundation that ties into new services and deployment models such as the Cloud. As part of the transition, for new services the Cloud may be a no-brainer as a computing model, but for existing shared services one has to conduct an assessment to determine if the Cloud provides value. If applications are not designed for the Cloud or does not leverage the capabilities that the Cloud has to offer this may be an issue. New Cloud services have to tie into the service foundation and with existing services that have already been built.
I highly recommend Enterprise maps that lay out the services foundation and their interactions. I have worked on the development of such maps in the past and they can be an extremely useful tool to specify and map services. Existing services can be at the Enterprise or Department levels and can include infrastructure, application and business levels. Many organizations have an Enterprise service bus can facilitate the communications between the software services and applications. The bus supports Enterprise application integration with functions such as communications, monitoring, deployment controls, event handling and mapping. For existing and projected services at these levels an evaluation has to be done to come up with migration candidates to the Cloud and the associated timelines.
In addition, decisions need to be made whether the services will be private, public or community based. Community services foster high levels of reuse since multiple tenants can be organizations with similar needs such as government agencies, health care, finance etc. The infrastructure services can map to the Cloud infrastructure as a service or a conglomeration of these can map to platform as a service such as frameworks that used to build applications. Application and business services can map to Cloud software as a service. As part of the transition, data portability is important and the data should be able to be easily processed and parsed when moved to the Cloud. Updates for user management may have to be made related to directory services.
The design of the applications would also need to be developed or updated to leverage the on demand and elastic capabilities of the Cloud. Security may need to be beefed up due to the multi-tenancy aspects of the Cloud. Monitoring elements will need to be defined, such as data that needs to be monitored and potential automated scaling triggered by such data. In addition, infrastructure updates such as reduction in the number of servers, consolidation of data centers and server virtualization can be part of the process. Application bandwidth requirements should be evaluated so that latency is not an issue after the migration. Any such bottlenecks that are identified, should be addressed to facilitate adequate response times. Building end-to-end Cloud solutions as next generation shared services will require a special focus on further developing the service foundation and addressing design considerations for Cloud migration.
• • •
(This has been extracted from and is a reference to Ajay Budhraja's blog).