Three tools that governments can use to modernise legacy systems

Improving user experience lies at the heart of much of the work put into modernising legacy systems, according to a report on Govtech.com, which highlights user experience as a key motivation behind current modernisation efforts by governments.

This is because legacy systems typically carry technical debt that can slow down organisations, including government agencies. This can lead to frustrated users, costs and lost opportunities.

Data platform Redis’ Senior Solution Architect, Raden Ardhian, shares three key features of an application modernisation solution that enables governments to transform their legacy IT into modern, agile, and scalable systems.

1. A data communication layer to enable microservices

Contrary to building a monolithic structure, organisations can opt to build microservices, a software architecture strategy that breaks down apps into loosely coupled services that are independently deployable.

As legacy systems tend to struggle with data silos, a rich data communication layer can enable an organisation to modernise a legacy system as a microservices architecture.

This layer enables information to be stored in a structured and standardised way, such as JavaScript Object Notification (JSON) or Hash fields, and allows users to access the relevant data for their respective purposes, says Ardhian.

For instance, Redis’ data integration capabilities, including Redis streams, allow data from multiple sources to be integrated in real-time, which frees up time for developers and enables them to focus on the larger business strategy, he adds.

“Redis streams are known for being high performance with low latency, and plays a major role in providing efficient inter-service asynchronous messaging within microservices,” Ardhian explains.

With microservices, governments can easily scale innovations at speed, respond to citizen demands, and launch new products faster.

2. Rich data structures to meet diverse needs

Government apps have wide-ranging needs – from managing citizen data to tracking other systemic issues.

Data structures are a means of organising and storing data in computers, and have diverse use cases across different profiles. Therefore, it is key for data structures to deliver flexible ways to organise data for modern apps, says Ardhian.

Ardhian shares that Redis makes data structures available in both unstructured and structured forms. These include deterministic and probabilistic capabilities, as well as geolocation capabilities.

For example, individual citizen profile information can easily be managed with structured data, while tracking holistic nationwide statistics can be optimised with probabilistic data structures, such as Hyperloglog or Redis Bloom filter.

Probabilistic data structures use algorithms to estimate solutions for large-scale problems, rendering them useful for nationwide policymaking.

“We have existing Redis government projects that have been leveraging Redis for over five years, scaling user-session and profile data use-cases that could not have been possible without Redis Enterprise” he explains.

3. A vector database for generative AI (GenAI) applications

Unlike a traditional database, a vector database is designed to efficiently store complex data types like images, videos and audio – making such databases particularly suited for AI-driven data analysis.

Cost savings, performance and security are some critical considerations when organisations design GenAI solutions for high-scale requirements such as in government agencies, says Ardhian.

Redis vector databases have been optimised to process a high volume of data with minimal delay, and remove the requirements of training large language models (LLMs) by exposing confidential information, he adds.

“Most government projects start with security. [For example], rail stations and airports need real-time processing of fraud detection scenarios, such as face detection, payment processing, ticketing,” he says.

Redis vector database also incorporates retrieval augmented generation (RAG) technology, which can improve the accuracy of LLMs by retrieving responses only from approved data sources.

Other features include semantic caching and chat history built-ins made available to its clients within its RedisVL library that enables developers to easily add the functionalities in a chatbot application.

Redis’ Senior Solution Architect, Raden Ardhian, will be speaking at AWS’ Public Sector Day Indonesia on 10th July 2024, in Jakarta, on accelerating enterprise application modernization in government agencies. You can find more information about the event here >>>