NYC++: January 2024 at Tower Research ft., BlazingMQ

Come to our January meetup to kick off 2024! Tower Research Capital will be hosting us in their downtown offices, with food provided by NYC++.

Please fill out this form before RSVPing. Registrants who have not filled out the form will not be able to attend and will removed from the RSVP list in the days leading up to the event.

Featured Speakers: Vincent Yan and Taylor Foxhall
Talk Title: Building Low-Latency Large Scale Message Queues in Safe C++

Description
This talk will provide a practical breakdown of the techniques applied to build BlazingMQ, a large-scale distributed queueing system designed to support enterprise applications. The goals were clear: build low-latency reliable infrastructure that could support highly concurrent network traffic, and do it using pre-modern C++. More specifically, the highly concurrent and asynchronous nature of message queues pushed us to consider early on how to architect this system for safety.

Nearly a decade after it was architected, built, and deployed, we published BlazingMQ as open source. So, now's a good time to ask, how did we try to stay in the utopic realm of safe C++? Was there anything we tried that was particularly effective, and can we recommend ways that you too can stay safe in C++?

For BlazingMQ, this meant abstracting cross-thread communication, leveraging contract programming, tracking memory across various application subsystems, and of course, testing extensively. These design decisions have consistently protected BlazingMQ from bugs in the usual categories that plague C++ applications, such as memory safety, thread safety, and even logical safety.

Speaker Bios
Vincent Yan is a software engineer at Bloomberg. He is particularly interested in distributed systems and computer networks. As a core contributor to BlazingMQ, he employs safe C++ practices and various testing techniques to ensure the system's robustness and performance. Vincent earned a bachelor's degree in coomputer engineering and a master's degree in computer science from Johns Hopkins University.

Taylor Foxhall is a software engineer at Bloomberg. They have a particular interest in systems programming and the design and evolution of programming languages, and are focused on building better and safer APIs to make C++ easier to use for everyone. Taylor eaerned a bachelor's degree in computer science from Binghamton University.