Building Infrastructure for a Trillion Daily Transactions
Speaker: Tony Savor, Vice President, Platform Engineering, Index Exchange
Date: Tuesday, January 13, 2026
Time: 11:00 a.m. – 12:00 p.m.
Location: In-person at 700 University Avenue, 9th floor, Room 9016, Toronto and online
Abstract:
What does it take to run software that handles trillions of requests every day—without slowing down or breaking? At Index Exchange, our platform powers one of the world’s largest real-time digital marketplaces, where milliseconds matter and “rare” failures occur routinely at global scale. In this talk, we’ll explore how owning and operating our infrastructure enables us to scale safely, efficiently, and predictably. Drawing on real production experience, we’ll examine challenges such as rewriting a live system in Go without downtime, designing architectures that remain resilient under constant failure, and using profiling, metrics, and experimentation to uncover hidden performance bottlenecks. Along the way, we’ll share broader lessons about distributed systems at scale, including why tail latency (p99 and beyond) shapes user experience and business outcomes, and how edge cases become the norm as systems grow. Whether you are a student, working in a fast-paced technology environment, or engaged in research adjacent to large-scale systems, this session offers eye-opening and highly relevant insights into how modern infrastructure is built and operated in practice. From real-world challenges to the design decisions that keep the Internet’s real-time economies moving, this talk will leave you with a deeper understanding of what it actually takes to build resilient, high-performance systems at planetary scale.
Speaker Biography:
Tony Savor has been leading engineering and operations of mission critical, planet scale infrastructure for over 30 years. Most recently, he led Google Kubernetes Engine (GKE) Engineering, defining the next generation distributed computing infrastructure. Kubernetes is used by thousands of companies and the second largest open source project in the world (only behind Linux). In the past, he led engineering of Facebook's online data storage and serving infrastructure. Serving the social graph at tens of billions queries per second, this was the largest system of its kind in the world. Prior, he led engineering of a foreign-exchange retail/institutional trading platform at OANDA.com, trading 10's of Billions of dollars per day. Tony holds a PhD in Computer Engineering from the University of Waterloo and is an Adjunct Professor at the University of Toronto.
Speaker’s Online Presence: