Add a blog post for LIKE optimizations #8576
Conversation
✅ Deploy Preview for meta-velox ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
facebook-github-bot
added
the
CLA Signed
|
cc @mbasmanova |
| relaxed patterns that are not so straightforward: | ||
|
|
||
| - `hello_velox%`: matches inputs that start with 'hello', followed by any character, then followed by 'velox'. | ||
| - `hello_velox%`: matches inputs that end with 'hello', followed by any character, then followed by 'velox'. |
There was a problem hiding this comment.
typo: hello_velox% -> %hello_velox ?
There was a problem hiding this comment.
Nice catch! Fixed.
xumingming
force-pushed
the
like_opt_blog
branch
from
91cbba7
to
b18effe
Compare
There was a problem hiding this comment.
@xumingming James, thank you for writing this nice. It reads very well.
@pedroerp Pedro, would you also take a look?
|
@mbasmanova has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
@mbasmanova has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
There was a problem hiding this comment.
Thanks @xumingming for this blog post. It is very useful.
Had some nits in the writing.
|
|
||
| ## What is LIKE? | ||
|
|
||
| <a href="https://prestodb.io/docs/current/functions/comparison.html#like">LIKE</a> is a very useful operation, |
There was a problem hiding this comment.
Nit: Use multiple short sentences instead of commas.
"LIKE is a very useful SQL operator. It is used to do string pattern matching. The following examples for LIKE usage are from the Presto doc:"
|
|
||
| - Use `%` to match zero or more characters. | ||
| - Use `_` to match exactly one character. | ||
| - If we need to match `%` and `_` literally, we can specify escape char to escape them. |
There was a problem hiding this comment.
Nit : specify "an" escape character
| into Velox's function call, e.g. `name LIKE '%b%'` is translated to | ||
| `like(name, '%b%')`. Internally Velox converts the pattern string into a regular | ||
| expression and then uses regular expression library <a href="https://github.com/google/re2">RE2</a> | ||
| to do the pattern matching. RE2 is a very good regular expression library, it is fast |
There was a problem hiding this comment.
Same here : Use full stop between the sentences "RE2 is a very good regular expression library. It is fast and safe, which gives Velox LIKE function a good performance."
| expression and then uses regular expression library <a href="https://github.com/google/re2">RE2</a> | ||
| to do the pattern matching. RE2 is a very good regular expression library, it is fast | ||
| and safe which gives Velox LIKE a good performance. But some popularly used simple patterns | ||
| can be optimized to use simple C++ string functions to implement directly, |
There was a problem hiding this comment.
Nit : "can be optimized using direct simple C++ string functions instead of regex."
| and safe which gives Velox LIKE a good performance. But some popularly used simple patterns | ||
| can be optimized to use simple C++ string functions to implement directly, | ||
| e.g. Pattern `hello%` matches inputs that start with `hello`, which can be implemented by | ||
| memory comparing the prefix bytes of inputs: |
There was a problem hiding this comment.
Nit : " can be implemented by direct memory comparison of prefix ('hello' in this case) bytes of input"
| Although these patterns look similar to previous ones, but they are not so straightforward | ||
| to optimize, `_` here matches any single character, we can not simply use memory comparison to | ||
| do the matching. And if user's input is not pure ASCII, `_` might match more than one byte which | ||
| makes the implementation even more complex. And also note that the patterns above are just for |
There was a problem hiding this comment.
Nit : "Also note that the above patterns are just for illustrative purposes. Actual patterns in practice can be more complex."
| } | ||
| ``` | ||
|
|
||
| Here `cursor` is the index in the input we are trying to match, `unicodeCharLength` is |
There was a problem hiding this comment.
Maybe format this as follows
Here :
- 'cursor' is the index in the input we are trying to match.
- 'unicodeCharLength' ....
So the logic is basically repeatedly....
| a function which wraps utf8proc function to determine how many bytes current character consists of, | ||
| so the logic is basically repeatedly calculate size of current character and skip it. | ||
|
|
||
| It seems not that complex, but we should note that this logic is not effective for pure ASCII input, |
| so the logic is basically repeatedly calculate size of current character and skip it. | ||
|
|
||
| It seems not that complex, but we should note that this logic is not effective for pure ASCII input, | ||
| for pure ASCII input, every character is one byte, to match a sequence of `_`, we don't need to |
There was a problem hiding this comment.
Nit sentence :
"Every character is one byte in pure ASCII input. So to match a sequence of '', we don't need to calculate the size of each character and compare in a for-loop. Infact, we don't need to explicitly match '' for pure ASCII input as all. We can use the following logic instead:"
| } | ||
| ``` | ||
|
|
||
| It only matches the kLiteralString pattern at the right position of the inputs, `_` is automatically |
There was a problem hiding this comment.
End this sentence with full-stop.
xumingming
force-pushed
the
like_opt_blog
branch
from
4a3769d
to
9d2dfd8
Compare
|
@aditi-pandit Thanks for the review, made corresponding changes to all the comments. |
|
@mbasmanova has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
@mbasmanova merged this pull request in 3004e34. |
|
Conbench analyzed the 1 benchmark run on commit There were no benchmark performance regressions. 🎉 The full Conbench report has more details. |
Summary: Pull Request resolved: facebookincubator#8576 Reviewed By: Yuhta, kgpai Differential Revision: D53308906 Pulled By: mbasmanova fbshipit-source-id: 31a1efe0d5472ccc9f2a1c81602e402d2f8c8e8a
No description provided.