Good morning everyone. Welcome to Con 42 IOT Conference. My name is Ilpa plo. I have over 20 years of experience in software development and more than 10 years, 10 years of experience in payments industry. Today I would like to talk about how microservices architecture is transforming payment systems for iot devices. Okay. This topic sits at the intersection of payments, distributor systems and connected devices, and it's becoming increasingly relevant as more devices begin to act on our behalf. To understand why this matters, let's first let look at what's changing in the payments landscape. We are in the middle of an iot payment revolution. Payments are no longer initiated, only by people. Devices themselves are making decisions and triggering transactions, sometimes thousands of times a day. Think about smart meters reporting energy usage, vehicles paying tolls, automatically appliances, recording supplies, traditional payment systems, assumed human interaction, infrequent transactions, and relatively large amounts. I iot completely breaks those assumptions. So what happens when we try to support this new world using old architectures? Most legacy payment platforms are monolithic, and that becomes a serious problems with iot. First, scalability. Vertical scaling simply doesn't work when thousands of devices suddenly come on our network. Second deployment, speed waiting weeks to deploy a security fix is unacceptable when devices are already deployed in the field. And third, technology locking older systems struggle to adopt modern cryptography, machine learning, or blockchain based integrations. And unfortunately, these aren't just performance issues. They introduce real operational risk. In monolithic systems, a failure in one area can quickly cascade across the entire platform. For iot use cases like connected vehicles or CD infrastructure, that kind of outage is simply not tolerable. Databases are another major bottleneck. They were never designed to handle real-time telemetry, high volume micro transactions, fraud signals, settlement records all at the same time. So Monolithics are in the answer. What is This is where microservices architecture come in instead of one large system. We break the platform into smaller independent services, each focused on a single business capability, authentication, authorization, fraud detection, settlement. Each becomes its own service with its own lifecycle. Let's look at why this architectural shift works as so well for IO OT payments, there are four core principles that make microservices a strong fit here. First, single responsibility. Each service does one thing and does it well. Second, independent deployment. We can update one service without risking the rest of the system. Third, technology flexibility. Each service uses the best tool for its job. And finally, fault isolation. A failure in analytics shouldn't stop payments from going through. Now let's walk through how this looks in practice. Starting with device authentication, IOT devices must authenticate on their own often with limited processing power, unreliable connectivity. This requires. Cryptographic credentials embedded during manufacturing. Secure elements for tamper resistance. Full cycle management, including ownership changes and decommissioning microservices allow us to manage the road, manage, and rotate credentials without physically asking the device. Once the device is authenticated, the next step is dev, deciding whether a transition should be allowed. Transaction auth authorization must happen extremely fast, often in milliseconds. We verify balances, subscriptions and usage limits. We access risk in real time. We validate geographic location to detect impossible scenarios, and we enforce item potency to prevent duplicate charges. All of this happens automatically at scale and without human involvement. But there is another challenge unique to iot transaction size. Many iot transactions are tiny fractions of ascent processing. Each one individually isn't part practical, so we aggregate them. For example, a smart meter may generate thousands of usage records per day. These are accumulated. Aggregated based on threshold or schedules, and then processed as a single payment. At the same time, detailed logs are preserved for transparency. After authorization and aggregation, we move into the most regulated part of the system. Settlement and fraud detection. Settlement is where money actually moves and it requires strong consistency and compliance. Microservices support multi-state settlement flows, continuous reconciliation and integration with banking networks. Running alongside settlement is fraud detection, powered by machine learning. We analyze device behavior, detect coordinated attacks, and apply risk-based controls without slowing normal transactions. To make all of this efficient, we rely heavily on a synchronous communication. In an event, written architecture services communicate through events instead of direct calls. When a device authenticates an event is published, fraud detection consumes it, analytics streams, it, notifications react to it. Platforms like Kafka allow us to scale this to millions of events per day. Perfect for iot. Of course, all these devices still need a secure endpoint entry point into the system. Ivo devices speak many protocols. MQTT co-op HGDP and automotive protocols. API Gateway translates these protocols enforces security, applies rate, limits, routes, requests to the right services. This protects backend systems from spikes, malformed requests and attacks. Let's, let's see how all of this comes together in a real world. In a smart city scenario, thousands of connected parking meters, vehicles and charging stations generate payments automatically. Revenue must be distributed across multiple agency, each with different rules and. Microservices allow each concern to be handled independently while still coordinating seamlessly, while still coordinating seamlessly. Running a system like this requires strong operational discipline. We rely on distributed tracing to follow requests across services. Metrics and monitoring helps us spot trends and anonymous anomalies. Intelligent alerting tells us when human intervention is required without observability, micro visas don't scale with it. They thrive. And none of this works without strong security foundations. We use multiple layers of security. Secrets management, service authentication, data tokenization, network segmentation, and defense. In depth, this approach meets strict financial regulations while supporting the scale and diversity of iot. So where does this leave us? Microservices provide a foundation that's scalable. Resilient unactive. They allow payment platforms to grow with the explosion of connected devices without sacrificing reliability or security. As billions more devices come online, architectures like, like this will be essential. So as a summary, as we have seen today, iot is redefining how payments happen and microservices provide the foundation to support that shift securely and at scale. This approach allows payments platform to grow, adapt, and continue innovating at the connected as the connected worlds expand. This is it from me today. Thank you for listening.