From one parent to another
When Provantage was engaged by Buck and Hickman to help with the migration of there user services from Farnell to BSS, we were delighted to help. The problem lay as usual, with a customer aversion to the new parent. Identity with the business with the new parent could drive away custom, so it had to be handled very carefully.
The businesses internal networks were also needing careful management. We didn’t want the new parent being able to find its way across into the old (and rival) parent. It got very complicated. The needs of both the old and new parent had to be taken into account at all stages of the migration, as well as keeping users working and accessing the old and new systems as data moved across.
Breaking down the problem
External identity, data, applications; three areas we had to be very careful on. Firewall rules, security policies, and new user policies were drawn up to ensure while they had access to the data they needed, they were prevented from accessing the data and systems they shouldn’t.
Email was a big problem and we came up with a novel solution that kept external email flowing as users were migrated across. Customers were able to retain the old email addresses, thereby giving continuity of service. Even after the migration had been completed, these email addresses could continue to be used.
On the user network, we created a way of having both the new and old parent companies on the same network devices, simplifying network migration.
In the meantime, moving the businesses catalogue data of some 30,000 lines had to be completed. A team of data engineers was drafted in to extract, transform, and load data into the new parent company’s systems. We were able to provide assistance in this area with code to help translate from one schema to another.
Monitor and react
During the migration process, each migrated store and office was constantly monitored for performance. All operations were carried out remotely with as small a physical change made as possible. During this process, the environment was tested and monitored closely for any out-of-band activity.
Due to the careful planning and monitoring we detected very few unusual events and even discovered poor network configurations which were causing some strange problems.
Users detected very little change, since we levelled up the policies on both sides so the only change was a different desktop background colour and image.
Upon completion of the project, we had migrated all the catalogue lines, upgraded fifty desktop applications, moved over 1,000 user accounts together with their desktops, and integrated the email service into one seamless capability.
All that was left to do was tear down the temporary networks, change some firewall rules, and disconnect some temporary servers. Using configuration as much as possible was one of our principles, with using temporary hardware second, ensured disconnecting the company from its old parent was as simple as running some scripts.