Bank of England Smoothly Migrates to Cloud

This is a closely governed, extremely risk averse organisation whose reputation are top priority. They have high-level auditing controls and run primarily on a traditional data centre model.

40 %

Hosting cost reduction

1 day

Developer onboarding time

50 : 1

Devs : DevOps ratio

Fast Facts

Analysis that used to take this organisation 20 days to accomplish can now be completed in half an hour.

We assisted this organisation by implementing four key factors: a secure landing zone, an on-boarding mechanism for applications, containerisation and training.

The challenge

The organisation had a pressing, time-sensitive need to port their legacy applications to the cloud. Some applications were in an on-premise data centre that needed to be vacated, whilst others were running on soon-to-be obsolete versions. Simply extending the existing data centre contract was not a viable option, with prohibitive costs and little to no return. The better solution was to leverage the cloud.

With a full understanding of the positive opportunities of cloud and how migrating these services would provide the catalyst to enable many of their other services to exploit cloud, they decided this was the right direction. They were motivated by meeting a long-term, technical, strategic objective, but also needed to reduce cost and increase resilience and availability of services.

It soon became apparent that although a migration might be relatively straight forward in a technical sense, there may still many sizeable organisational considerations to address to ensure full security of services, governed, supported and consumable by the end user, post transition.

The cloud holds great promise as a solution, although a transition of this scale may also encounter various issues. The organisation posed questions such as; How should we go about making this transition?, What are the best practices?, Which applications should be ported first and why?, How can we be certain the applications were truly secure in this new working environment?

These questions are only natural when making such a large shift. Unlike most cloud transition efforts, this organisation simply could not afford to learn lessons the hard way. The migration had to be done right, the first time, with no risk.

The solution

Appvia had already proven itself as a trusted technical delivery partner within the organisation on a previous project. With this in mind, teams approached us to strategise and execute their cloud migration, amend governance processes and introduce best-practices. We needed to also train, up-skill and transfer this best-practice knowledge to in-house teams to ensure continual delivery, post-migration.

The Appvia team consisted of engineering and delivery professionals, embedded into the organisation to collaborate with cyber security, service management, architecture and finance. Together they designed, built and implemented the following in Microsoft Azure:

A secure landing zone
Providing a single location for user interaction and a solid foundation for the
proper use of cloud technology. A secure landing zone allows for extensions and future growth.

An on-boarding mechanism for applications
Rather than each application having its own approach to build and deployment, we implemented a common CI/CD process. This technique ensures security will be incorporated into the applications as early as possible in the build process

Containerisation
Having introduced a new application development paradigm for cloud, we further refined it to include containerisation.

Training
We provided several courses in various topics including Git, Kubernetes and Helm. The training consisted of instructor-led classroom sessions complemented by self-serve exercises for delegates to complete at their own pace.

The results

This organisation is still early in their cloud adoption efforts, but their cloud adoption strategy is now mature, future-proof and based on solid security guidelines and best practices for Microsoft Azure.

Our journey with this customer didn’t end there, we are now helping them to realise their full potential with other processes. For example, we are working with them to implement a scalable, secure way to master Machine Learning, using Kubeflow and Azure batch. Analysis that used to take 20 days to accomplish can now be completed in half an hour.