Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Vignette on web server options in R #5

Open
JosiahParry opened this issue Jul 7, 2023 · 3 comments
Open

Vignette on web server options in R #5

JosiahParry opened this issue Jul 7, 2023 · 3 comments

Comments

@JosiahParry
Copy link
Owner

As suggested by @eddelbuettel, document REST / web server alternatives in the R space.

Notably RestRserve is a complete replacement for plumber & valve itself. Personal opinion is that plumber is less performant (probably) but has a lot of niceties such as generating open API docs automatically, etc.

Plumber is built ontop of httpuv which can be used as an alternative

Rook last commit was 2017 not sure if its still in use?

@JosiahParry
Copy link
Owner Author

OpenCPU is the other that I couldnt remember

@eddelbuettel
Copy link

We may be talking slightly past each other, and if so this may well be my fault for not groking what you are going about but to me the situation has nothing to do with 'web servers' or 'serving files from R' but more simply with providing REST entry points to (for example) provide an easy and fast predict() method given a simple covariate vector.

plumbr made this easy and possible. This came at a cost, the ease of use means it is not all that fast / performant for thousands of requests.

RestRserve solved precisely that problem. It is used in production, at scale, to server ML models.

valve appears to focus on being better at what plumbr does. By ignoring RestRserve in your otherwise compelling salesmanship you are providing an incomplete picture of the overall situation. I, and I suppose others, would love to see a three-way comparison between plumbr, RestRserve and valve.

@eddelbuettel
Copy link

eddelbuettel commented Jul 9, 2023

And then there are other endeavours like the (now-defunct) Syberia project (more at https://github.com/syberia/) which came out of local fintech that (as early as a decade) very much served R-based ML prediction models (of the classic classifier type) at scale.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants