Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Lookup benchmark #1738

Merged
merged 7 commits into from
Feb 26, 2020
Merged

Lookup benchmark #1738

merged 7 commits into from
Feb 26, 2020

Conversation

bright-starry-sky
Copy link
Contributor

Lookup benchmark , test result as below :

System info :

  • CPU : Intel(R) Core(TM) i7-9750H CPU @ 2.60GHz
  • MemTotal : 8G

Total vertices : 1000000

  • PreciseScan_10000 : Precise scan , return 10000 vertices
  • PreciseScan_100 : Precise scan , return 100 vertices
  • PreciseScan_10 : Precise scan , return 10 vertices
  • FilterScan_10 : Full index scan, return 1 vertex
====================================================
StorageLookupBenchmark.cpprelative  time/iter  iters/s
====================================================
PreciseScan_10000                   577.73ms     1.73
PreciseScan_100                    594.96ms     1.68
PreciseScan_10                      469.44ms     2.13
FilterScan_10                          1.57s  636.61m
=====================================================

@bright-starry-sky bright-starry-sky added the ready-for-testing PR: ready for the CI test label Feb 7, 2020
@bright-starry-sky
Copy link
Contributor Author

System Info

  • CPU : Intel(R) Core(TM) i7-9750H CPU @ 2.60GHz
  • MemTotal : 8G

Test result

  • Total vertices : 1000000
    • PreciseScan_10000 : Precise scan , return 10000 vertices
    • PreciseScan_100 : Precise scan , return 100 vertices
    • PreciseScan_10 : Precise scan , return 10 vertices
    • FilterScan_10 : Full index scan, return 1 vertex
=======================================================
src/storage/test/StorageLookupBenchmark.cpprelative  time/iter  iters/s 

=======================================================
PreciseScan_10000                                          31.81ms     4.12
PreciseScan_100                                            33.98ms     4.63
PreciseScan_10                                             28.96ms     4.52
FilterScan_10                                              260.16ms    2.82
=======================================================

dangleptr
dangleptr reviewed Feb 19, 2020
View reviewed changes
src/storage/test/StorageLookupBenchmark.cpp
* FilterScan_10 : Full index scan, return 1 vertex
**/

/**
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What's the index and expression?
What's the part number?

Copy link
Contributor Author

@bright-starry-sky bright-starry-sky Feb 19, 2020 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What's the index and expression?
What's the part number?

  • Part number :
    1 part

  • tag、index structure as below :

tag (col0 string, col1 string, col2 string)
index on tag(col0, col1, col2)
  • expression :
PreciseScan : col0 == xxxx
FilterScan : col1 == xxxx

The FilterScan does not work now. I need to improve this benchmark at later.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fine. Could you update the comments

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fine. Could you update the comments

Sure, comments updated.

@bright-starry-sky
Copy link
Contributor Author

To improve the benchmark.
Newly test result as below. Performance bottleneck in encode and decode when index full scan.

  • CPU : Intel(R) Core(TM) i7-9750H CPU @ 2.60GHz
  • MemTotal : 8G
  • Total vertices : 1000000
  • Total partition number : 1
  • tag (col_0 int, col_1 int, col_2 int)
  • index on tag (col_0, col_1, col_2)
  • PreciseScan_10000 : Precise scan , return 10000 vertices , where clause col_1 == 1
  • PreciseScan_100 : Precise scan , return 100 vertices , where clause col_1 == 1
  • PreciseScan_10 : Precise scan , return 10 vertices , where clause col_1 == 1
  • FilterScan_10 : Full index scan, return 1 vertex , where clause col_1 < 1
=====================================================
 src/storage/test/StorageLookupBenchmark.cpprelative  time/iter  iters/s
 =====================================================
 PreciseScan_10000                                          9.65ms   103.65
 PreciseScan_100                                            5.23ms   191.03
 PreciseScan_10                                             4.14ms   241.64
 FilterScan_10                                              1.47s  679.35m
 =====================================================

panda-sheep
panda-sheep approved these changes Feb 25, 2020
View reviewed changes
Copy link
Contributor

@panda-sheep panda-sheep left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM.

@dangleptr dangleptr merged commit 41fe7c8 into vesoft-inc:master Feb 26, 2020
@bright-starry-sky bright-starry-sky deleted the lookup_benchmark branch February 27, 2020 06:12
jude-zhu pushed a commit to jude-zhu/nebula that referenced this pull request Mar 4, 2020
@bright-starry-sky @dangleptr @jude-zhu
Lookup benchmark (vesoft-inc#1738)
f6407b2 
* Lookup benchmark

* To optimize the lookup benckmark

* all compact before index can

* improve benchmark

* Updated comments

Co-authored-by: dangleptr <37216992+dangleptr@users.noreply.github.com>
tong-hao pushed a commit to tong-hao/nebula that referenced this pull request Jun 1, 2021
@bright-starry-sky @dangleptr
Lookup benchmark (vesoft-inc#1738)
b5eaa50 
* Lookup benchmark

* To optimize the lookup benckmark

* all compact before index can

* improve benchmark

* Updated comments

Co-authored-by: dangleptr <37216992+dangleptr@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@dangleptr dangleptr dangleptr approved these changes

@panda-sheep panda-sheep panda-sheep approved these changes

@critical27 critical27 Awaiting requested review from critical27

@darionyaphet darionyaphet Awaiting requested review from darionyaphet

@dutor dutor Awaiting requested review from dutor

@laura-ding laura-ding Awaiting requested review from laura-ding

@liuyu85cn liuyu85cn Awaiting requested review from liuyu85cn

@sherman-the-tank sherman-the-tank Awaiting requested review from sherman-the-tank

Assignees
No one assigned
Labels
ready-for-testing PR: ready for the CI test
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
3 participants
@bright-starry-sky @dangleptr @panda-sheep