WIP: doc how to find out the bottleneck resources for trouble shouting high latency issue

[st1page]

Info

• Description

  • [ What's changed? Which parts of the docs are affected? ]

• Notes

  • [ Include any supplementary context or references here. ]

• Related code PR

  • [ Provide a link to the relevant code PR here, if applicable. ]

• Related doc issue

  Resolves [ Provide a link to the relevant doc issue here, if applicable. ]

For reviewers

• Preview

  • [ Paste the preview link to the updated page(s) here. Edit this item after the preview site is ready. To find the updated pages, scroll down to locate and open the Amplify preview link and select the dev version of the documentation. ]

• Key points

  • [ Parts that may need revision or extra consideration. ]

Before merging

• I have checked the doc site preview, and the updated parts look good.

• I have acquired the approval from the owner (and optionally the reviewers) of the code PR and at least one tech writer (CharlieSYH, emile-00, & hengm3467).

[aws-amplify-us-east-1]

This pull request is automatically being deployed by Amplify Hosting (learn more).

Access this pull request here: https://pr-1927.d2fbku9n2b6wde.amplifyapp.com

[fuyufjh]

LGTM 👍. It would be even better if we can put some example screenshots to illustrate what does the normal and abnormal cases look like.

[neverchanje]

IMO the key metric should be the backpressure rate.
If I understand correctly, when it's "slow" and hits some bottleneck, there must be some "backpressured" streaming jobs.
So the first step to identify the job should be identifying the jobs with highest BP value.

[st1page]

Yes, it is contained in the previous chapter "## Diagnosis —— find out the bottleneck streaming job" and has published in our doc