Migrating Large Enterprise Customers to VOSS-4-UC
Tim Jalland, Senior Proposition Manager, VOSS
Following on from this week's launch of the fantastic M2UC, I wanted to share some insights into a recent migration project from an early adopter customer who chose to use M2UC.
Heavy Duty Lifting
When it came to upgrading to a new cutting edge telephony and UC solution, it was no surprise that a large UK government customer came to VOSS to manage the process. The incumbent system was public-facing and provided critical business communications to customers, suppliers and over 100,000 employees across 1000+ sites.
The approach taken was to absorb the existing configuration and design, transform as required, and then migrate in-situ on the new system – a VOSS approach known as Synchronization and Overbuild – rather than build a new / separate system.
The process had to conform to some strict rules:
- The system was very much in use - so change windows must be kept to days not weeks
- The whole migration process had to be repeatable, robust and accurate (automated)
- The roll-back had to be practical, tested and well communicated
- Existing numbering plans and feature usage (such as phone settings, services, hunt groups) had to be transitioned fully to minimize the impact to users of the service
- No service data was allowed to leave the data center, and access had to be RBAC-protected
- Disruptive changes had to be kept to a minimum and always made out of hours
The underlying system was a multi-cluster Cisco Call Manager (5 x clusters) with an over-arching SME cluster (providing PSTN breakout and access to Unity voicemail) and managed from Cisco CUCDM8. The ask was to migrate this to VOSS-4-UC whilst also continuing to provide telephone services without a blip.
Six Important Points
With an estate of this size, a "big bang" approach was not appropriate (too much risk) and a staged approach was followed. Six important points were agreed to, to ensure all teams were aligned and the project – which was run on a tight timescale (3 months) – was successful.
- Assign model offices (test sites) and practise: the preparation, the migration, the roll-back, the training, and the configuration. Check and practise everything. Then check it again.
- Start with a single live pilot site – significant enough to identify areas that need further tuning, and allow sufficient running-in time for the site. Design inter-calling between old and new dial plans. Collate feedback and then retest in the model office.
- Transition cluster by cluster; more complicated in that each cluster in turn will migrate to the new system but more manageable in terms of resources and risk. Across each cluster, keep the change window to under 5 days for all common MACDs.
- Automate as far as possible – manual efforts rely on certain individuals, allow numerous errors to creep in, humans become tired at the end of overnight change windows and manual operations are not reliably repeatable. Automation and AI tooling such as M2UC from VOSS form the foundation for a successful migration of this scale.
- Test pre-migration and post-migration and count the number of end points registered – to validate everything that was working beforehand is also still working after the migration.
- Accommodate change – the project will change as it moves through but keep the end goal in view and expect and accommodate changes as part of the weekly task planner.
VOSS Overbuild Migration Architecture
The outcome? This was another successful migration for VOSS, with the customer on a solid foundation for the future.
To find out more about VOSS Migration Services and the VOSS M2UC Migration Tool, please contact us.