GopherCon 2021: Kaylyn Gibilterra - Go at Mainframe Scale53:11 36 views 100% Published a month ago
At Capital One we have created a distributed system in Go to process all account data from any modern core or legacy mainframe across the company, with peak traffic of up to 2 million tps and 5Gbps, storing over 20TB on different storage systems across multiple regions in AWS.
We had not established widely adopted Go enterprise service standards, so this required on-the-job library creation, scaling challenges, ramping up an organization of engineers new to Go, designing and redesigning microservice boundaries, all based on hiring Kaylyn as the single Go SME to stand it up. This was a bold decision. Fortunately for this project, it produced great results.
In this talk, Kaylyn will cover our biggest challenges and successes to give a practical guide to how any company can transition to writing Go services, from any language. Thanks to Go’s simplicity and scalability, we could linearly scale our teams and exponentially scale our services - and it’s far easier than you might imagine.