News for hackers. Stuff that matters.
Re:The State of Serverless GPUs by Pulcinella on Saturday April 29, 2023 @06:44PM (#35757991)
I literally don’t know what it means. Does serverless just mean doing it locally? Like an on premises cluster of GPUs in the IT closet?

If you are doing it on someone else computer, and that computer is a big server farm, I find it odd to use the term “serverless.” And by odd I mean literally the exact opposite of serverless.

Re:The State of Serverless GPUs by throwaway70701 on Saturday April 29, 2023 @08:37PM (#35758670)
"Serverless" is a cloud-native software development paradigm. Of course there is a server - the machine itself, and there is a HTTP server running on the machine. The point is in the software you write.

A classic "serverful" backend would expose a HTTP server on a port and handle client connections.

Whereas a "serverless" program is invoked by a server on-demand instead of running all the time, and it doesn't expose anything on any port. The client request is passed to it (e.g. on standard input) and it's supposed to pass the response to the server (e.g. through standard output) - and the server sends it to the client.

Do you know PHP? That's the original serverless. Compare that to writing a Django app in Python, or Express/Node.js, or ASP.net. As opposed to these, your page.php is invoked only when someone opens that page by the Apache daemon.

The point of serverless is to forget (from developer's POV) about servers as in software as well as servers as in machines.

Re:The State of Serverless GPUs by matt_heimer on Sunday April 30, 2023 @01:30AM (#35759909)
Having a server is like having a car. Its yours, you do what you want with it but:

1) If you don't drive much then you might be overpaying

2) You have to worry about the car infrastructure, you have to maintain it and if it breaks down you might be offline (unable to travel) until its repaired.

3) If you need bring home a washer and dryer for example you have to worry about your car having the needed capacity (scaling)

In todays world you can decide not to own a car and just use Uber or whatever vehicle sharing service is available to you. You pay only for the rides you need, you don't have to worry about infrastructure (repairs) and you can always pay for larger vehicles on demand.

That's serverless computing. Instead of paying for a physical computer or a virtual computer (compute service) you deploy the app you want to run and pay for each run of your app. Typically the app is some type of request/response or event processing app. The cloud company maintains the compute fleet and when a request comes in the will deploy your app to one of their compute nodes long enough to service the request.

Its called serverless not because there are no servers but because you don't know or care about the server it runs on. The hardware could change out between executions. You don't care about failed hardware, OS patching, etc. The lowest level you typically deal with is selecting the app language (node, java, etc).