SerializationString improve performance

[kitaisreal]

Changelog category (leave one):

• Performance Improvement

Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):

Improve performance of string serialization.

[kitaisreal]

Prepare table:

CREATE TABLE test_table (id UInt64, value String) ENGINE=ReplacingMergeTree ORDER BY (id) SETTINGS index_granularity = 32768;
INSERT INTO test_table SELECT number, number % 15000 FROM numbers(1500000000);

Before:

SELECT max(length(value))
FROM test_table
FORMAT `Null`

0 rows in set. Elapsed: 0.855 sec. Processed 1.50 billion rows, 19.89 GB (1.75 billion rows/s., 23.27 GB/s.)
Peak memory usage: 1.24 MiB.

After:

SELECT max(length(value))
FROM test_table
FORMAT `Null`

0 rows in set. Elapsed: 0.691 sec. Processed 1.50 billion rows, 19.89 GB (2.17 billion rows/s., 28.79 GB/s.)
Peak memory usage: 1.17 MiB.

[robot-ch-test-poll3]

This is an automated comment for commit 6cab012 with description of existing statuses. It's updated for the latest CI running

⏳ Click here to open a full report in a separate page

Successful checks

Check name	Description	Status
Mergeable Check	Checks if all other necessary checks are successful	✅ success

Check name	Description	Status
CI running	A meta-check that indicates the running CI. Normally, it's in success or pending state. The failed status indicates some problems with the PR	⏳ pending

[alexey-milovidov]

Please add the changelog entry and merge with master.

[alexey-milovidov]

Looks like a clear win!