What a week it’s been! The ExCeL London was buzzing with energy as nearly 2,000 data engineers, architects, and decision-makers gathered for Current 2025, Confluent’s flagship event dedicated to data streaming.
Lightstreamer was proud to be there as an exhibitor, contributing to the engaging conversations around real-time architectures, Kafka ecosystems, and the last-mile delivery of live data. It was an opportunity not just to showcase our technology—but to deeply connect with the community driving the future of streaming.
Real-time, right to the front end
As the conference explored the growing adoption of diskless architectures, observability tools, and mission-critical use cases for Kafka, we found ourselves in many conversations centered on one core theme: how to get real-time data all the way to the user’s screen—fast, reliably, and securely.
That’s where Lightstreamer comes in. Our Kafka Connector and rich set of client libraries enable organizations to deliver live data from Kafka to web, mobile, and desktop applications with unmatched efficiency and flexibility.
Whether it was CTOs looking to modernize their legacy infrastructure or developers searching for robust real-time APIs, the interest in Lightstreamer’s “last-mile” capabilities was strong and steady throughout the event.
A hub of connections
Our booth welcomed a remarkable stream of visitors—both new faces and long-time contacts. We reconnected with existing customers, caught up with partners, and met a wide array of Kafka practitioners who shared insights, questions, and challenges.
It was a pleasure to hear firsthand how Lightstreamer is being used across industries—from financial platforms requiring ultra-low latency, to media companies ensuring live updates reach global audiences instantly.
Highlights from the floor
Here are a few key takeaways we observed at the event:
- Diskless is hot: Operator interest soared in Kafka’s diskless topics—where brokers write directly to object storage instead of local disks—driven by benefits like eliminating cross‑AZ replication costs, simplifying scaling (no more segment replication), and reducing infrastructure failures tied to disk management. Data flows through an in‑memory buffer, then asynchronously lands in S3‑style storage, enabling broker restarts in seconds and seamless stateless scaling. While write latency increases slightly (100–400 ms at P99), this architecture shines in high‑throughput or cost‑sensitive scenarios.
- Observability is top-of-mind: With Kafka lag, throughput variance, and resource usage becoming critical concerns, teams are doubling down on observability tools to gain real-time visibility into their streaming infrastructure. Popular solutions include Grafana, Prometheus, and OpenTelemetry, among others. These tools are being used to trace message flows end-to-end, monitor broker health, identify consumer bottlenecks, and proactively resolve issues before they affect SLAs.
- The ecosystem is evolving: We saw renewed interest in hybrid streaming architectures, where Lightstreamer acts as the bridge between Kafka backends and highly dynamic frontends.
Looking ahead
Current London 2025 made it clear: streaming is not a trend—it’s the foundation of modern digital systems. And while Kafka powers the backbone, Lightstreamer ensures that data reaches the end user with precision, resilience, and scalability.
To everyone who stopped by our booth: thank you! We’re excited to continue the conversations—and to help more organizations harness the full potential of their real-time data.
Want to learn more or see a demo? Get in touch with our team. We’d love to show you what’s possible.
Giuseppe Corti
Alessandro Alinone
Alessandro Carioni
