FPS achieves its mission through key partnerships with institutions like Sigedis, NISSE, and CBSS, ensuring effective pension management, maximizing citizen benefits, and reinforcing Belgium’s social security system.

“This vision underscores our commitment to key values and drivers: customer orientation, process automation and simplification, and a human-centric organizational approach.” Tuur Bressink, Application Architect SFPD

A Deep Dive into FPS’s Applications

The FPS application landscape is extensive, headlined by mypension.be, a user-friendly portal where citizens can manage and simulate their pensions. Key applications include:

• Theseos: The main pension application, originally for employee pensions.
• Satellite Applications: Emerging applications gradually becoming independent from Theseos.
• PensionPortal and ApSoft: Incorporated from the merger with PDOS for civil servant pension calculations.
• PCP (Pension Communications Platform): Facilitates pension-related communications.
• Specialized Applications: Various tools for specific tasks, integrated post-mergers.

This diverse range of applications shows FPS’s commitment to efficient and adaptable pension management.

Architectural Innovation at FPS: Pioneering a New IT Framework

At the Federal Pension Service (FPS), a broad application landscape and a forward-thinking mentality go hand in hand. However, FPS were facing a set of challenges.

An aging workforce, not fully replaced, and difficulties in attracting and retaining IT talent posed significant resource challenges. Aging applications and legacy systems, some of which were inherited through mergers, were adding to the complexity of the landscape, causing technical debt. The Theseos system, in particular, exemplifies this issue. What started as a solution for employee pensions had grown into a cumbersome monolith with more than 300 modules, making it unwieldy and difficult to manage. This complexity impacted development productivity, as seen in extended feedback cycles where code compilation and CI pipeline processes were notably time-consuming and inefficient.

To tackle its challenges, FPS launched Project Renaissance. A strategic move to shift from a monolithic system architecture to a suite of domain-oriented applications. This transition is designed to improve observability within the distributed IT ecosystem, leading to more efficient and effective service delivery.

The move towards a modular application landscape means that each application now has specific, well-defined responsibilities. In this distributed setup, the ability to gather relevant insights is more crucial than ever, underlining the importance of applied observability.

To enable this, FPS established essential architectural patterns to guide the behavior and interaction of these applications, ensuring a streamlined and cohesive operational environment.

Supporting this new architecture, FPS has rolled out several key platforms:

• Axway API Gateway for API management, ensuring that applications can communicate effectively.
• A Service bus based on Apache Camel (Redhat Fuse) allows for lightweight integration flows when needed.
• A Message bus built on Redhat AMQ caters to use-cases that don’t require immediate feedback.
• A central IAM system, utilizing Redhat SSO (Keycloak), to streamline identity and access management across all applications.

These platforms generate a vast array of logs, events, and metrics. To fully leverage this data, FPS identified the need for an observability stack. The goals for implementing this stack are clear:

• To be proactive in identifying and resolving problems early, which allows FPS to concentrate on adding value.
• To enhance the quality of service provided to the public.
• To implement a ‘Shift Left’ approach—integrating metrics not just in production but also during development, testing, and user acceptance testing (UAT).

Why FPS is Embracing Applied Observability

To manage the substantial data generated by their observability platforms, we assisted FPS in designing and implementing a robust applied observability strategy.

This applied observability strategy marks a shift from traditional, reactive monitoring based on historical data to a proactive engagement with data. Applied Observability (AO) focuses on responding to current information and taking immediate action. Similar to how real-time data navigation apps like Waze operate, proactively addressing issues as they arise. This approach involves ingesting data, analyzing trends, and preempting potential problems.

By focusing on the specific issue  of non-targeted and non-actionable information, this approach effectively tackles common IT challenges related to logging and monitoring. By doing so, it avoids situations where crucial alerts are overlooked because they’re lost in a sea of irrelevant data.

“We’re shifting from relying on historical KPIs and moving towards analyzing real-time data, recognizing patterns, and responding to the present. Instead of predicting the future based on the past, we’re equipping our Observability Platform to make decisions in the now. Incorporating AI for decision-making is our next leap in maturity — stay tuned for that evolution.” Matthias De Scheerder, Integration Architect and Managing Partner at Archers.

Observability starts with monitoring individual components, where we gather basic data to understand each element of our system. As we progress, we achieve structured observability, which allows for greater data consolidation, giving us a comprehensive view. Next is causal observability, leveraging a time-series topology to understand the relationships and sequences of events over time. Finally, we advance to proactive observability with AIOps, automating root cause analysis, business impact analysis, and alert correlation for a predictive and preemptive IT strategy. More on that in the following blog.

The aim of applied observability at FPS is multifaceted. It scales productivity and customer value by advancing solutions and product delivery. It pioneers customer engagement with accelerated responses, pursues sustainable technology solutions, and optimizes resilience and operations.

Implementing Applied Observability at SFPD

The implementation of Applied Observability at SFPD was strategically centered on the Archers connectivity framework. This framework served as a critical enabler for embedding the initiative within the organization.

Business Strategy

There are three business drivers at the core of the FPS applied observability strategy:

• Job Enablement: Equipping staff with the right tools for their jobs.
• Customer Satisfaction: Ensuring users of services, from middleware to APIs, are satisfied. This extends beyond FPS to the end-users.
• Protecting Investment: Securing the platforms’ integration into the observability framework to ensure that applications are stable en perform well. For instance, if FPS invests in an IAM tool, it’s imperative that the platform remains stable to avoid financial loss.

Organization

The vision and strategy of the Federal Pension Service (FPS) are centered around creating an observability framework that extends through every layer of the organization. The approach is comprehensive, starting from the bottom-up, ensuring that service providers are integrated into the solution and process their own events. This enables consumers, the end-users, to monitor services effectively.

“Our initial focus is on the essential infrastructure, network and middleware layers. These are vital for the smooth operation of applications and processes. In 2024, our goal is to strengthen these areas as well as advance to application layers”, Matthias De Scheerder, MP Archers

The team driving this initiative includes IT architects, platform specialists, analysts, and developers, forming an Observability Enabling Team. This team collaborates with service providers – those handling critical organizational functions like network and security – to ensure seamless integration with the observability platform.

Service Providers within the organization, such as the Core Business Infrastructure (CBI) team, Integration team, and specific product teams, are clearly defined and self-organized to deliver their services successfully. They are responsible for implementing observability in their domains, encompassing visualization, alerting, and communicating critical events to consumers.

A centrally managed Enabling Team leads in exploring and adopting innovative observability tools and practices. They provide guidance, establish best practices, and offer coaching across the organization.

Key users are identified at all organizational levels to play a crucial role in achieving observability. The long-term goal is to transition observability responsibilities from the Center of Excellence to these key players, ensuring a sustainable and effective approach.

Platform

To support the observability project, this step focuses on validating and establishing a platform architecture. This involves a thorough assessment of the different platform architectures and their landscapes. This assessment provides a clear overview of all platforms and the specific capabilities they serve. Additionally, aligning these architectures with the reference framework helps define clear responsibilities and scopes. This alignment also identifies overlaps, leading to proposals for simplifications in the overall landscape.

For technical insights into application operations, observability tools like Grafana and Prometheus are selected. For the functional insights, Dynatrace is in place to tackle this.

Design and Realization

Design and realization involve leading the exploration and adoption of new observability technologies. with an emphasis on establishing best practices and providing appropriate recommendations. This includes integrating data meaningfully into the platform, correlating data, and creating guidelines around visualization and alerting. Collecting requirements, understanding what data needs to be gathered from different assets, and defining event messages are all part of the process. Once this information is gathered, it is systematically organized and structured in Jira for efficient implementation.

Governance

To streamline the Service Delivery Life Cycle (SDLC), a governance checklist will be developed for use by various self-organized Service Providers within the organization. This checklist aims to provide a foundational set of guidelines to ensure successful service delivery.

Additionally, these governance guidelines will facilitate new service providers in integrating with the observability framework and effectively managing new requirements within their services.

Marketing and Evangelization

Marketing and evangelization efforts position observability as a strategic topic for 2024 and beyond. Communicating the ambitions, business drivers, and strategy continuously and making progress visible within the organization is key. Initiatives like technical open spaces are leveraged to bring focus to the project.