Closed timfee closed 1 year ago
can you post the link here?
Update from @emrysal
I think that's a bad ticket - it's been improving slowly ever since that was assigned to me but I'm not sure what is possible and it's continual improvement
We also experience this issue. A lot. @Jaibles
we have an ongoing task to improve performance of this. the main issue is that the cold boot of serverless functions is slow. this gets better if your app is used more frequenlty -- i know not a perfect solution. we're trying to find a way to do this without cold boot
note to self: https://cal.com/peer/30min is faster, surprisingly
What API is it using? What file? Random guess: https://github.com/calcom/cal.com/blob/main/apps/web/pages/api/book/event.ts
You might consider using Edge API Routes, which will help you get faster cold startups: https://nextjs.org/docs/api-routes/edge-api-routes
export const config = {
runtime: 'edge',
}
export default async function handler(req: Request) {
return new Response()
}
But I think would need to verify this is possible with next-auth
.
note to self: https://cal.com/peer/30min is faster, surprisingly
the getSchedule
request for your calendar takes almost 3s for me to complete as well
~ β― time curl 'https://cal.com/api/trpc/viewer.public.slots.getSchedule?batch=1&input=%7B%220%22%3A%7B%22json%22%3A%7B%22eventTypeId%22%3A127%2C%22eventTypeSlug%22%3A%2230min%22%2C%22usernameList%22%3A%5B%22peer%22%5D%2C%22startTime%22%3A%222023-01-02T23%3A00%3A00.000Z%22%2C%22endTime%22%3A%222023-01-31T22%3A59%3A59.999Z%22%2C%22timeZone%22%3A%22Europe%2FBerlin%22%2C%22duration%22%3Anull%7D%2C%22meta%22%3A%7B%22values%22%3A%7B%22duration%22%3A%5B%22undefined%22%5D%7D%7D%7D%7D' > /dev/null --compressed
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1394 0 1394 0 0 684 0 --:--:-- 0:00:02 --:--:-- 686
curl --compressed > /dev/null 0.02s user 0.01s system 1% cpu 3.048 total
Woah.
In this release, we have decreased the size of our engine files by an average of 50%. The size of the Query Engine used on Debian, with OpenSSL 3.0.x, for example, went from 39MB to 14MB. We will also remove some smaller engines to decrease the total size in future versions.
https://github.com/prisma/prisma/releases/tag/4.8.0
You might want to update! (Assuming this API route uses Prisma)
@lukasklein
the getSchedule request for your calendar takes almost 3s for me to complete as well
well 3 seconds is still better than 5 to 10 seconds (which we have seen too)
anyone interested in this, we're making a bit of progress. stay tuned.π
we have identified that one of the slowness reasons is a slow caldav server for the affected users.
we can still improve prisma and cold boots tho
/bounty $1000 per second
π $1,000.00 bounty created by PeerRich
π To claim this bounty, submit a pull request that includes the text /claim #4535
somewhere in its body
π To receive payouts, join Algora and complete the relevant onboarding steps
π΅ Payment arrives in your account 2-5 days after the bounty is rewarded
π― You keep 100% of the bounty award
π³ If you want, you can donate 100% of the rewards to climate change projects!
π Thank you for contributing to calcom/cal.com!
@PeerRich I have an idea that could solve this, i will submit a pull request soon.
π‘ roae submitted a pull request that claims the bounty π To accept and/or reward this claim, visit your org dashboard on Algora
@roae cal wants to reward your solution. Please sign up on Algora, link your Github account and connect with Stripe on your dashboard.
roae has completed their onboarding. You can now complete the payment on your org dashboard
ππ @roae has been awarded $1,000! ππ
we ended up hiring @roae π
@timfee we just released Data Proxy. Can you verify if it's still slow?
See attachment. Still seems like it's a bit slow, but better than before!
https://user-images.githubusercontent.com/3246342/225504906-9a1148e1-7913-4d4a-a92e-381b31428fd0.MOV
See image -- on a fast connection, with a single calendar configuration -- painfully slow.