London Borough of Hackney
London Borough of Hackney
I Need Help Service [COVID-19 help service for vulnerable residents]
Briefly describe the initiative/ project/service; please include your aims and objectives
The tool was created in response to an overwhelming request for help from residents, particularly vulnerable residents of the borough when we went into lockdown as a response to the coronavirus pandemic. This included categories such as shielded residents – over 70s, pregnant women and residents with health issues. To respond to this need promptly we leveraged several AWS services and Platform APIs to quickly develop and deploy a solution, with the first version going live within 4days. Over 1,200 households in Hackney utilized this service to ask for urgent help with food & medicine, mainly because they couldn’t leave the house and their normal support mechanisms weren’t available. This project used Platform APIs which are reusable data domain driven APIs to reduce development time and cost for existing and future services that will need data from a given data domain. Furthermore, the project also focuses on building a reliable cloud infrastructure, with data migration from on-premises to a cloud environment, thus making data and service availability greatly increased, aiming to reach an availability of 99.99999%. The project also includes the development of an API Hub that will be used for easy discovery of available APIs.
What are the key achievements?
– Reduced development time & cost
– Great improvement on data availability and thus, services availability
– Less dependency on on premise data sources.
– Team confidence in using cloud products and developing solutions including Solutions Architect certifications within the team.
– Secure, highly available and reliable cloud environment.
– Reusable data pipeline from on premise product to cloud based environment.
– Reusable API template (boilerplate) with all tools and dependencies already set up, which greatly reduces development time of new APIs and ensures consistency across.
– Improved CI/CD pipeline, deploying to separate AWS accounts, which provides better separation of environments.
– Stable and scalable architecture.
– APIs not built on demand or for random services, but driven by existing data domains, which reduces dependency on backend development and speeds up the delivery of new services.
– Custom authorizer, which provides better security and fine access control down to the API endpoint level.
– Provides confidence that users will only be able to consume an API endpoint they should have access to.
– Catalogue of available APIs (API Hub), to ensure easy discoverability of new and existing APIs and prevents possible duplication of work.
What are the key learning points?
- Built our core skills including our apprentices.
- Ensuring we have a consistent and reusable code base for all future APIs via our boilerplate Base API.
- Have well documented solutions to easily onboard new developers and simplify maintenance and support.
- Have a responsive support mechanism to respond to application issues fast.
- Collaboration with various teams including Security team, Information & Governance and Data & insight team for PIA and API usage framework.
- Data migration methods can vary massively depending on use case – built 3 ways of migrating data to account for possibly known scenarios
- Introduced the usage of Lambda, replacing existing ECS with EC2/Fargate set up as it was better suited for our use case.
- Templated infrastructure set up using infrastructure as code – All AWS resources are now created using Terraform and Serverless framework, allowing for quick creation of resources and easier maintenance. As most of our APIs AWS infrastructure setups are the same, we produced templates to reduce development time and make provisioning of all required
- resources happen in minutes.
- Producing API documentation prior to development using Swaggerhub
COVID-19 Response Recognition Award
I Need Help service [Covid-19 help service for vulnerable residents] – The tool was created in response to an overwhelming request for help from residents, particularly vulnerable residents of the borough when we went into lockdown as a response to the coronavirus pandemic. This included categories such as shielded residents – over 70s, pregnant women and residents with health issues. In order to respond to this need in a timely manner we leveraged a number of AWS services and Platform APIs to quickly develop and deploy a solution, with the first version going live within 4 days. Over 1,200 households in Hackney utilized this service to ask for urgent help with food & medicine, mainly because they couldn’t leave the house and their normal support mechanisms weren’t available. This project used platform APIs which are reusable data domain driven APIs. By creating ‘Platform APIs’, we reduce development time and cost for existing and future services that will need data from a given data domain. Furthermore, the project also focuses on building a reliable cloud infrastructure, with data migration from on-premises to a cloud environment, thus making data & service availability greatly increased, aiming to reach 99.99% availability.