This is something I’ve been trying to figure out on my end. I run a community instance for creators at https://spectra.video, and so far it’s been a pretty good situation. The community is fairly cozy, we don’t have many problems with spam, and we connect with a handful of really good partner instances that we like and trust.

Here’s the thing: my server keeps growing.

In like the last few months, my server’s user count has ballooned up to 320 users, from about 50 or so. That’s in spite of the fact that I’m incredibly conservative about user registration, and screen just about everyone who registers to cut down on trolls and spam.

Object Storage is sitting at half a terrabyte for video, which in the grand scheme of things is actually not terrible. But, the server is not beefy for constant video transcoding, and setting up infra to make that pain easier is not cheap. At all.

At the end of the day, I’m going to have to think about migrating to a more powerful VPS, to setting up runners on some kind of cloud GPU architecture, to thinking about whether to segment out my database to some kind of externally hosted solution. All of this stuff is expensive, migration is messy, and trying to support all of this seems complicated.

I feel like, within the next few months, we’re going to see an influx of YouTube migrations similar to what happened with Reddit and Twitter. I don’t think any particular instance can truly be ready for that to happen, but I’d like to at least try to prepare for the worst.

Is anybody else thinking about how to deal with this?

  • thegiddystitcher@lemm.ee
    link
    fedilink
    arrow-up
    3
    ·
    edit-2
    1 year ago

    It might be worth getting in touch with the admin of MakerTube.net (if you haven’t already) as we’ve had a similar influx there. They’ve posted updates letting us know everything is ok for now, but that it might not be if we keep getting this YouTube migration happening.

    I’ve already offered to help with non-technical stuff there but the tech side is over my head, sorry. Hopefully if PT admins work together like a lot of the Lemmy admins do there’s more chance of a plan that works for everyone.

  • spaduf@slrpnk.net
    link
    fedilink
    arrow-up
    3
    ·
    1 year ago

    I feel like the solution to this problem will vary significantly from instance to instance. Who’re your primary users (by resource utilization) and are they in a good position to chip in?

  • Jeena@jemmy.jeena.net
    link
    fedilink
    arrow-up
    9
    arrow-down
    1
    ·
    1 year ago

    I feel we PeerTube admins should at least try to share the runners to utilize them as much as possible. Nowadays they are mostly idling because each instance doesn’t really get constant uploads.

      • Sean Tilley@lemmy.mlOP
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 year ago

        Hey, Chocobozz!

        Yes, I’m aware of the runners, but I do not have the expertise or the infrastructure to set them up effectively. Part of my feedback is: there isn’t much guidance on how to do runners correctly, and setting up runners with something like cloud GPUs or even lots of little render nodes is prohibitively expensive and/or time consuming.

        It may be my logical next step, but I kind of wish we had some community infrastructure for runners, to make this less painful.