DATA CENTRE TRANSFORMATION A CASE STUDY

dct01

The exponential evolution of IT hardware platforms, operating systems and applications has created a quasi-perpetual data centre transformation trend.

Every company, regardless the size is undergoing the data centre transformation process from hardware and applications refresh, to virtualization and cloud migration. If your company has its IT infrastructure migrated to cloud, the burden of hardware refresh/upgrade will be carried on by the cloud provider, your IT department will only look after the applications upgrades.

The case study refers to a typical medium size company who requested a Data Centre Transformation (DCT) project with migration to new data centres and an assessment of what kind of transformation will be suitable for their IT infrastructure at the new data centres.

The approach was a typical enterprise architecture one (loosely following the TOGAF mainframe).

We organized the first workshop with the client’s management, stakeholders and IT management and technical people.

We learned that the customer will take care of the business transformation and also of the data/applications migration and transformation. Our role was to perform the IT infrastructure transformation and migration.

The subsequent workshops were dedicated to understand the business needs, directions and growth for the near and distant future and also to get a similar understanding of the company data structure and growth, data protection and archiving needs, also the applications matrix, future requirements, virtualization cloud compatibility and mobility requirements.

The results of these workshops was that we learned that the growth of the business will be on average of 5% per annum, the data will grow on a rate of 10% per annum, a dedicated backup and archiving solution was required (we were able to provide a leveraged secure backup solution but it was not accepted), the applications will not be transformed prior or during migration, but virtualized after a successful migration project, and most important the customer required zero downtime for its Production environment during migration and this requirement was not negotiable!

The team tasked with the IT hardware platform migration consisted of a Project Manager (PM), an Enterprise/Solution Architect (SA), a Business Analyst (BA), a Network Engineer (NE), two Windows/VMware Engineers (WVE), a Storage SME (SE) and a Backup SME (BS).

We started with a Discovery phase, auditing the Prod and DR data centres, recording every equipment, compute, storage, network and security.

The results of the Discovery phase were presented in a report and a workshop with all the relevant stakeholders. The customer IT infrastructure consisted of two data centres, a Prod live one and a cold DR one. The Prod infrastructure had new servers with growth capacity, needed a new SAN and a refresh of the core switches and encryptors. The Prod environment was setup with redundancy in the Prod data centre, each server had an equivalent server on standby.

The cold DR data centre was outdated, with old servers and network equipment. The fail over process was fully manual and very slow, totally unacceptable for the business.

The solution for the customer DCT was the following:

  • Migrate the cold DR infrastructure from the existing cold DR data centre to a new DR data centre and transform it into a live DR data centre acquiring an active-active data centres scenario. The new active DR data centre will be fitted with new servers and network equipment (similar or newer to the Prod environment), the fail over process will be 90% automated and scripted with minimum manual input (only for the core applications requiring multi-level authentication).
  • In order to achieve the zero downtime requirement for the Prod data centre migration/transformation, a staged process was designed: First stage, a fail over of half of the Prod applications from the old Prod data centre to the new active DR datacentre and migrate the infrastructure belonging to those applications to the new Prod data centre. After migration another fail over process was performed, from the DR data centre to the new Prod data centre. Second stage, failover of the remaining applications from the old Prod data centre to the new DR data centre. Migrate the remaining infrastructure from the old Prod data centre to the new Prod data centre and perform another fail over from the DR data centre to the new Prod data centre.
  • The design was catering for the applications interdependency; the staged migration of the Prod data centre was based on moving dependent applications during the same stage.
  • The design provided a new solution for the Prod network infrastructure with new core switches and encryptors and also a redundant Internet connection sourced from three major Telcos.
  • The design also provided a new backup solution, with new backup servers and a dedicated backup network.
  • Although it was not part of the migration/transformation solution, the customer requested a solution design for a virtualization of their Prod and DR environments, new and more performant SAN and a high level design about a possible cloud migration, which we provided.

The migration/transformation of our customer IT infrastructure was a success, the main requirement of zero downtime during migration was achieved together with a streamlined Prod environment, ready for virtualization and cloud migration.

If you want to find out more about DCT projects and maybe want some help with your migration/transformation contact me: ilie.mihut@dashinternational.com.au