Oleg Lobanov is a Staff Engineer at Bitvavo with 16 years of experience in engineering. He leads the architectural vision and technical evolution of Bitvavo’s ultra-low-latency exchange platform, drawing on deep expertise in building distributed, fault-tolerant, and latency-critical systems.
Oleg specializes in advanced performance optimization, including JVM garbage collection minimization, efficient inter-process communication, and highly optimized data structures.
Achieving single-digit microsecond latencies in crypto trading isn’t just ambitious—it’s transformative. Founded in 2018 in Amsterdam, Bitvavo has grown into the largest Euro-denominated crypto exchange, setting benchmarks for performance, security, and scalability.
As one of the first exchanges licensed under MiCA, Bitvavo upholds unmatched transparency and trust across Europe. Recently, its engineering team rebuilt the core architecture on JVM technologies, unlocking single-digit microsecond latencies, boosting reliability, and creating a scalable foundation for rapid growth.
Key challenges addressed included ultra-low latency, concurrency, resilience, replayability, auditability, and data integrity. To meet them, Bitvavo applied advanced JVM optimizations: minimized garbage collection, extensive object pooling, efficient IPC via reliable UDP, high-performance ring buffers, and specialized data structures.
High availability was ensured through in-memory replication with hot, warm, and cold standby nodes, delivering near-zero downtime and robust resilience.
This talk offers practical insights into JVM performance tuning, architectural strategies for resilient high-throughput systems, and best practices for high availability—sharing lessons learned, pitfalls to avoid, and actionable guidance for building mission-critical, high-performance applications.
Searching for speaker images...