-
Notifications
You must be signed in to change notification settings - Fork 5.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add D1 changelog and release notes for request latency improvements #19235
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
1 files reviewed, 1 total issue(s) found.
Co-authored-by: hyperlint-ai[bot] <154288675+hyperlint-ai[bot]@users.noreply.github.com>
e4a09f2
to
377eb1f
Compare
377eb1f
to
5e06ad4
Compare
Co-authored-by: Matt Silverlock <[email protected]>
try to fix parse error
escape quote to fix yaml?
fix indentation
Deploying cloudflare-docs with Cloudflare Pages
|
@Oxyjun to approve/merge Friday his time |
date: 2025-01-07T18:00:00Z | ||
--- | ||
|
||
[D1](/d1/) lowered end-to-end request latency via the [Workers API](/d1/worker-api/) by up to 60% through eliminating redundant network round trips for each request. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is framed in terms of what we did
Vs
What does it now mean for you
Ex
When you query D1 from Workers, the round trip time is n% lower etc
|
||
This performance improvement benefits all D1 Worker API traffic, especially cross-region requests where network latency is an outsized latency factor. For example, a user in Europe talking to a database in North America. D1 [location hints](/d1/configuration/data-location/#provide-a-location-hint) can be used to influence the geographic location of a database. | ||
|
||
For more details on how D1 removed redundnat round trips, see the D1 specific changelog [entry](/d1/platform/changelog/#2025-01-07). |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Weird that I have to go somewhere else?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@irvinebroque I disagree. I want discoverability and navigation between changelog-next
(product updates) and D1 changelog
(release notes). A user skimming changelog-next
entries should take away what changed and why does it affect me, without necessarily caring about the how.
In an ideal future:
product changelog
would be generated release notes from code rollouts with more technical, implementation descriptionchangelog-next
would be product updates from code rollouts/shipped features that are worth calling out
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
your call! just saying that i initially interpreted the changelog post as, "whoa the query times are slow!" and then only after understanding "we got rid of network roundtrips" did it click that — "ohhh the total time it takes to get data back from D1 over the network is what is getting faster, we're not talking about timing of D1 queries themselves"
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
good feedback, let me try to clarify/make the improvement more clear without full details
description: |- | ||
D1 lowered end-to-end Worker API request latency by 40-60% by eliminating redundant network round trips for each request. | ||
|
||
For each request to a D1 database, at least two network round trips were eliminated. One round trip was due to a bug that is now fixed. The remaining removed round trips are due to avoiding creating a new TCP connection for each request when reaching out to the datacenter hosting the database. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Bringing this up into post helps me as reader understand
Ohhhh the metrics you're sharing are not query latency, they are total network time
Quick diagram may go long way
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I would also put the image in both sections.
addressing remaining comments, then will merge |
No description provided.