@keithwillcode But what would you recommend for someone who is in the middle of all of this? Honestly, if I have build something by ai and it breaks down, I feel even more frustrated than I used to before and I don't know the cure of it.
thinking of changing the way i post in x.
Instead of sharing the update,i should share how does the changes helps,what it improves,what it was lagging and what it improves
Any Ideas,Suggestion ?
Learning new new stuff daily is fun, but right now I really want to go back to basics and relearn computer networks, OS, and computer organization from scratch.
Part 13[Update]
- did added the test scripts for other 4 routes
- even though the latencys are under 1000ms in each of them but will try what i can do to improve them further.
- will be working on improving the latency on each of this routes
- yes leeetcode too because it's fun!
Part 12 [Update]
- downgraded my EC2 machine from t3.small to t3.micro, as I don't foresee needing t3.small in the near future.
- tested the POST request on EC2; the hot query takes about 975ms, so I think we're good.
- will be solving the leetcode afterwards because why not.
shards because that's how backend has been designed for each tier.
Using Promise.all to query all shards in parallel, not one by one.
Attaching the gist of after and before optimizations.
https://t.co/dVfZfGdJes
https://t.co/oJSjBnlcAE
review critique comment appriciated.
Part 11.2 [Optimizations]
We are using Idempotency key to ~
- Idempotency key stops double booking.
- shardKey given than request hits that one shard directly.
- No shardKey than request fans out to all shards.
Optimization
- send one query to the Primary DB, skipping empty π§΅