Part Two

For part one, click here.

In Part 1, we’ve identified the problem themes, now we’ll outline, at the high level, the different activities involved in our discovery, the value they bring, and what problem themes they relate to.

Whether you face one or all the issues mentioned above, it is important to state that discovery, by definition, is not intended to address all problems; it should not aim to capture completely exhaustive information, but find a balance between speed and thoroughness – focusing on the information that is critical to success.
Migrations are complex. Ecosystems are not only complex but also unique and constantly changing. The activities aim to not only understand these complexities, but also accommodate for their change.

Problem theme: Skills and maturity in cloud and DevOps


Activity: Arrange workshops and access to people, process and systems

The realignment of people across different levels of the organisation accelerates and ensures value is delivered.
By optimising the relationship between people, process and technology, teams are empowered to remove bottlenecks and inefficiencies that may be preventing your teams from working to their fullest.

Activity: Carry out a DevOps maturity and cloud readiness assessment

Client current state is examined against 19 performance indicators. These indicators align with industry best practice and are grounded in more than ten years of hands-on delivery experience (at the Ministry of Justice, Her Majesty’s Revenue and Customs, the Department for the Environment, Food, and Rural Affairs, Her Majesty’s Passport Office, the Met Office, and Lloyds Banking Group, among many others).

Problem theme: Envisioning what your migration journey should look like


Activity: Establish motivations and expectations

In the recent AL IT leaders survey, we found that loss of vision is one of the key barriers to success. We define and create the migration project charter, including communication plan, reporting and escalation procedures, roles and responsibilities, and ways of working between AL and the customer. As such, we can help you identify the areas most likely to impact your strategic objectives and challenges.

Problem themes: Risk of migration + Understanding of ecosystem


Activity: Discover and catalogue workloads and dependencies

Gather information on the in-scope workloads, including architecture, dependencies (internal and third-party), interactions, requirements (incl. hardware, software, storage, network, licences, security), and record in a Workload Migration Dossier. This should include (but isn’t necessarily limited to):

Business

  • Total cost of ownership (current plus estimate for migrated state, if possible)
  • Downtime acceptability
  • Regulatory/compliance requirements

Technical

  • Workload definition
  • Assessment of hardware capacity assigned to workload on-prem and whether this can be right-sized in the cloud to save cost (given the ease of scaling once migrated)
  • Metrics and KPIs, including importantly bandwidth/latency requirements to other apps/services/data stores, which may remain as-is
  • Connectivity requirements to/from internal and external systems (is ingress/egress from the public internet required?)

Activity: Capture each workload’s attributes in a Workload Migration Dossier

Per workload, this is the single source of truth that includes the Business and Technical information. WMD evolves over time as information is discovered and workloads are migrated to ensure that accuracy and clarity is consistently maintained.

Activity: Prioritisation and recommendations

For Business continuity and service resilience, an early view of workload migration priorities and migration readiness accelerate the Path-to-Value and ROI.

Case Studies


I’ve used the example of two engagements where we implemented our workload migration methodology with a successful discovery. The reason I’ve used these two specifically is because they emphasise how discovery is different depending on the need of the customer; the Met Office required a discovery geared more towards strategy and business alignment, whereas the Central Government client required a discovery to accelerate their implementation and time to value.

The Met Office

The Challenge

Faced with rising operating costs, the Met Office embarked on a major cloud transformation programme in 2018. Having attempted to deliver the same programme on two previous occasions without success, and with multiple third parties pulling in different directions and being guided by disparate strategies, a fundamental change was required.

How we solved it

We worked in partnership with the client’s team (as well as with a range of third-party providers, including AWS) to deliver a new, evidence-based strategy for this programme. This involved appraising the client’s capabilities, establishing a robust, DevOps-led delivery framework (underpinned by new, standardised processes), and co-defining the team’s charter, success criteria and ways of working.

Result

We added lasting value to the client by establishing a robust, evidence-based strategy and improvement framework that not only delivered a joined-up approach backed by a common vision between the third parties but also reduced the client’s costs.

Central Government: Security Sector

The Challenge

The client needed to lift-and-shift their on-premises estate to the Azure cloud, which included multiple production databases and applications. AL were engaged to support the transition to the Digital Service Operations team: to provide stable resources (including skilled engineers and a delivery manager), up-skill civil servants, improve existing services, reduce costs and introduce DevOps best practices where appropriate.

How we solved it

We started with an in-depth discovery phase to understand the underlying infrastructure, and then identified improvements to simplify the architecture. To improve the team and work towards a cultural shift, tasks were grouped into workstreams. We allocated mini-teams to work on each workstream comprising a mix of both AL engineers and Civil Servants at a variety of skill levels.

Result

By decommissioning unused resources, and purchasing Azure reservations for resources that had long term futures, we reduced both the client’s annual Azure bill by over 20% and the number of supported VMs from over 800 to under 600. Teams are now self sufficient and need half the number of people to support the workload.

Written by Danish Muhammad 

< Back