Making R run Fast - Jan Vitek

Channel:John O'Neill · Length:1:21:38 · Published:2016-03-02 · Views:870

Watch on YouTube

Summary

Jan Vitek of Northeastern University gives a meetup talk on the challenges of making the R language run fast. He explains what in R's semantics makes it hard to optimize and to build a JIT compiler for.

Why HN recommends it

HN points to it as the best explanation of why R is so hard to optimize, often paired with talks on Python and used to explain what makes Julia fast.

Another talk on the difficulty of optimising R is this fantastic one by Jan Vitek — HN, 2017

Software performance & optimization · Compilers, interpreters & program analysis · Data science & statistics

Subtitles (excerpt)

thank you uh vipa and IBM for hosting us uh tonight we have a a great show uh y VCH from Northeastern is going to tell us all about his work in making our run on well I think the llvm but but perhaps more General than that so Yan is actually fairly new in Boston just a year or so he's from Purdue uh his specialty is I guess compilers he's more of a computer scientist than a uh statistician but his wife I think is a statistician so um and let's see so I think you're on the board of sick plan or some on the board of sick plan so uh Kurt has uh so just very briefly next month or actually coming up in about two weeks we have another one two weeks we have Max um he's coming up he's talking about uh quintland cubus package rule based decision treat um I've SE before it's a good talk then we got to figure out what we're do for so if you want to talk about you know you have some ideas about what you want to see next please hang around after the talk and we can discuss that if you have some good ideas or you know someone you'd like to see we can try to arrange that uh so with that as an instruction uh let's see so uh restrooms are out and to the right uh I guess there's too many other logistics and so I'll I'll turn over to Yan right hello

On Hacker News

I recommend watching this video - Making R run fast [VIDEO] It's a little disappointing, because the conclusion is that R will probably never "run fast", but very interesting nonetheless. — lottin, 2018-10
Perhaps the best way to understand what makes Julia fast is to watch these two videos about Python and R and what makes them so hard to optimize: [LINK] [VIDEO] Take everything mentioned in these videos that make Python and R really hard to optimize and don't do those things :D — StefanKarpinski, 2018-06
Another talk on the difficulty of optimising R is this fantastic one by Jan Vitek: [VIDEO] Also, my favourite R trick: foo <- function() {rm(foo, envir=environment(foo))} — simonbyrne, 2017-07
This is disregarding the development of said ecosystems though. The point is that Python has been quite inhibitory to the development of this ecosystem. There are many corpses of automatic differentiation libraries (starting from autograd and tangent and then to things like theano to finally tensorflow and pytorch) and many corpses of JIT compilers and accelerators (Cython, Numba, pypy, and Tenso… — ChrisRackauckas, 2023-02

Related videos