Support approximate hypertable size

[nikkhils]

If a lot of chunks are involved then the current pl/pgsql function to compute the size of each chunk via a nested loop is pretty slow. Additionally, the current functionality makes a system call to get the file size on disk for each chunk everytime this function is called. That again slows things down. We now have an approximate function which is implemented in C to avoid the issues in the pl/pgsql function. Additionally, this function also uses per backend caching using the smgr layer to compute the approximate size cheaply. The PG cache invalidation clears off the cached size for a chunk when DML happens into it. That size cache is thus able to get the latest size in a matter of minutes. Also, due to the backend caching, any long running session will only fetch latest data for new or modified chunks and can use the cached data (which is calculated afresh the first time around) effectively for older chunks.

[codecov]

Codecov Report

Attention: 18 lines in your changes are missing coverage. Please review.

Comparison is base (d86346f) 79.80% compared to head (abe58e8) 79.83%.

Files	Patch %	Lines
src/utils.c	84.48%	7 Missing and 11 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #6463      +/-   ##
==========================================
+ Coverage   79.80%   79.83%   +0.02%     
==========================================
  Files         190      190              
  Lines       37148    37218      +70     
  Branches     9418     9427       +9     
==========================================
+ Hits        29646    29713      +67     
- Misses       3118     3122       +4     
+ Partials     4384     4383       -1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[fabriziomello]

One missing thing here is tests for OSM chunks (you can check test/sql/chunk_utils_internal.sql).

[nikkhils]

One missing thing here is tests for OSM chunks (you can check test/sql/chunk_utils_internal.sql).

Added in that file.

[mkindahl]

Looks good, but it would be good to test and ensure that it behaves sanely with a view or a foreign table (or other weird objects such as composite types or sequences) and also return NULL if the table does not exist (similar to how pg_relation_size works).

[mkindahl]

Since vacuum is needed to ensure that all forks exist, it might be a good idea to ensure that it works also when that is not the case.

[nikkhils]

Vacuum is only needed on PG13. I added this only for the compression case for PG13. In all other uses, it works without VACUUM. Newer versions pre-create all the forks

[mkindahl]

LGTM

[timescale-automation]

Automated backport to 2.13.x not done: cherry-pick failed.

Git status

HEAD detached at origin/2.13.x
You are currently cherry-picking commit 2b8f98c61.
  (fix conflicts and run "git cherry-pick --continue")
  (use "git cherry-pick --skip" to skip this patch)
  (use "git cherry-pick --abort" to cancel the cherry-pick operation)

Changes to be committed:
	new file:   .unreleased/pr_6463
	modified:   sql/size_utils.sql
	modified:   src/utils.c
	modified:   test/expected/size_utils.out
	modified:   test/sql/size_utils.sql
	modified:   tsl/test/expected/chunk_utils_compression.out
	modified:   tsl/test/expected/chunk_utils_internal.out
	modified:   tsl/test/shared/expected/extension.out
	modified:   tsl/test/sql/chunk_utils_compression.sql
	modified:   tsl/test/sql/chunk_utils_internal.sql

Unmerged paths:
  (use "git add <file>..." to mark resolution)
	both modified:   sql/updates/latest-dev.sql
	both modified:   sql/updates/reverse-dev.sql

Job log