Timely Dataflow in Rust

https://air.mozilla.org/bay-area-rust-meetup-may-2016/

Hello all!

I'm pleased to announce our Thursday May 12th meetup, with our special guest Frank McSherry who has been doing some fascinating work on incremental big data computations. Doors will open at 6pm, and talks will begin at 7pm. As always, Mozilla will be live streaming and providing food and drink. I'll see you soon!

Abstract: I'll introduce timely dataflow in Rust, a distributed programming framework introduced by Naiad, and implemented in a systems language we love.

Timely dataflow takes Rust's approach to zero-cost abstractions, providing a distributed programming experience where you only pay for what you use. Timely dataflow gets your code up and running in dataflow operators, and introduces coordination only where you explicitly request it. This allows fine-grained control when you want it, and a library of useful pre-fabricated operators when you don't.

We will talk through the timely dataflow stack, from the low-level primitives, up through libraries built on top (as examples), up to differential dataflow a high-level declarative programming language for incremental and iterative big data computation. Along the way, we'll try and pay attention to how Rust lets us do things without introducing performance-crippling abstractions. Bio:

Frank McSherry is an independent researcher formerly affiliated with Microsoft Research, Silicon Valley. While there he led the Naiad project, which introduced both differential and timely dataflow, and remains one of the top-performing big data platforms. He also works with differential privacy, due in part to its interesting relationship to data-parallel computation. Frank currently enjoys spending his time outside of Silicon Valley.

edit: air mozilla link (https://air.mozilla.org/bay-area-rust-meetup-may-2016/)