[Doc] improve document of query profile #58367
Conversation
|
@murphyatwork in "Structure of Query Profile" I found it at the end: Operator: This is a fundamental unit of execution in the Pipeline Engine. Each operator is responsible for implementing a specific algorithm, such as aggregation, join, or scan. So, since this definition is so important, I will find a spot for it at the beginning unless you think everyone will know this. I would like to copy this section to the top of the doc (still have it toward the end, but also at top so people have the background knowledge as they read the first half. The Pipeline Engine is a key component of the StarRocks execution engine. It is responsible for executing the query plan in a parallel and efficient manner. The Pipeline Engine is designed to handle complex query plans and large volumes of data, ensuring high performance and scalability.
Concepts in the Pipeline Engine:
Operator: This is a fundamental unit of execution in the Pipeline Engine. Each operator is responsible for implementing a specific algorithm, such as aggregation, join, or scan.
Pipeline: A pipeline is a sequence of connected Operator instances. It represents the execution path of a query plan.
PipelineDriver: A pipeline is instantiated as multiple drivers to support data-parallel execution. Each driver is responsible for executing a portion of the pipeline in parallel.
Schedule: Pipelines are scheduled in a non-blocking manner, utilizing user-space time-slicing. Operators are designed to be non-blocking and CPU-intensive, while other threads handle blocking operations such as disk I/O and network I/O. |
|
@murphyatwork I am in the middle of edits and moving the wide tables to Markdown structure that fits on the screen. I will finish tomorrow my time Good doc, I am learning a lot. |
Thank you for the suggestion. I'm also exploring ways to make it clearer. Perspectives from others are incredibly valuable. |
Signed-off-by: DanRoscigno <dan@roscigno.com>
Signed-off-by: DanRoscigno <dan@roscigno.com>
|
I pushed two commits @murphyatwork to 1) replace tables that were too wide 2) fix some spelling |
Signed-off-by: DanRoscigno <dan@roscigno.com>
|
Thanks so much for this @murphyatwork I learned a lot while reviewing it. I made some formatting changes, some of the tables were too wide so I converted them to lists. I also edited the other pages in the query profiles section. The file I removed, index.html, was causing problems unrelated to your PR (when I upgraded Docusaurus the other day it broke some functionality related to reviewing PRs, so I removed the file to review your PR) |
[Java-Extensions Incremental Coverage Report]✅ pass : 0 / 0 (0%) |
[FE Incremental Coverage Report]✅ pass : 0 / 0 (0%) |
[BE Incremental Coverage Report]✅ pass : 0 / 0 (0%) |
|
@Mergifyio backport branch-3.4 |
|
@Mergifyio backport branch-3.5 |
|
@Mergifyio backport branch-3.3 |
✅ Backports have been created
|
✅ Backports have been created
|
✅ Backports have been created
|
Signed-off-by: Murphy <mofei@starrocks.com> Signed-off-by: DanRoscigno <dan@roscigno.com> Co-authored-by: DanRoscigno <dan@roscigno.com> (cherry picked from commit 8a645dd)
Signed-off-by: Murphy <mofei@starrocks.com> Signed-off-by: DanRoscigno <dan@roscigno.com> Co-authored-by: DanRoscigno <dan@roscigno.com> (cherry picked from commit 8a645dd)
Signed-off-by: Murphy <mofei@starrocks.com> Signed-off-by: DanRoscigno <dan@roscigno.com> Co-authored-by: DanRoscigno <dan@roscigno.com> (cherry picked from commit 8a645dd)
Why I'm doing:
What I'm doing:
Fixes #issue
What type of PR is this:
Does this PR entail a change in behavior?
If yes, please specify the type of change:
Checklist:
Bugfix cherry-pick branch check: