Disaster tolerant solution ensures online services are available for customers who choose the internet to handle their tax and benefits.
The level of control we had through the suite of monitoring tools was extraordinary, as too was the level of collaboration between different people across different teams. Tim East, Deputy Director (Operations) IMS Live Services talking about the Self Assessment Online filing peak in January 2010
HM Revenue & Customs (HMRC) is making the internet the ‘channel of choice’ for customers, so new online tax services must be easy to use and always available. A reliable e-services infrastructure was needed to handle the growing number of users and to withstand problems as extreme as the total loss of a data centre. To underwrite this commitment, HMRC chose a disaster tolerant framework instead of more traditional disaster recovery, ensuring availability and meeting budget constraints. Working with IT partner Capgemini, HMRC’s new shared platform and advanced service management tools were successfully introduced for the new Self Assessment Online service. New VAT, corporation tax and PAYE web services are now being added.
The new shared platform is an infrastructure backbone and software architecture supported by sophisticated service management. Disaster tolerance is achieved by a dual data centre deployment, which is scalable and maximises use of existing assets. ‘Capacity on demand’ prioritises services during critical periods when user numbers peak. The software has been architected to securely manage and regulate user sessions across and within data centres, switching components on and off as required. During development, intensive tuning of the end-to-end solution included performance testing of cross government infrastructure components. Live service is monitored by system and service management capabilities that give real-time visibility of performance, availability and throughput for all system components, including external traffic dependencies. Service oriented principles guided the design and best practice processes are applied to ongoing management.
The architecture-driven solution is the backbone for the staged introduction of new web based services. Key benefits include:
- Web services are available when customers want to use them. If the infrastructure fails users are automatically diverted to working components and the segregation of services ensures near zero loss of data.
- Potential problems are identified early so preventative measures are taken before customers are impacted. Monitoring technology allows experts in mission control to proactively spot issues.
- Value for money from the decision to implement disaster tolerance instead of more costly solutions, to the ongoing operations of streamlined systems and processes.
- Cost savings from decommissioning and consolidating platforms when new web online services are migrated to the new shared infrastructure.
The new online services IT platform was delivered on time and under budget, supporting Self Assessment Online and a range of other critical services. It was introduced in phases to minimise risk, and resilience was put through its paces by the 50 percent growth in Self Assessment Online customers in its first year increasing to 6.5 million submissions in the 2009-10 filing period. During peak filing the service handled over twelve complex submissions a second.