You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In the metrics engine scenario, when the number of primary key columns on a physical table becomes excessively large, the CPU overhead for encoding all primary keys increases dramatically. This results in significant performance bottlenecks for both write and query operations.
Proposal
Introduce a sparse primary key codec that encodes only non-null primary key columns. By skipping the encoding of null or empty columns, this change aims to:
Reduce CPU overhead associated with primary key encoding.
Improve overall write and query performance.
TODO
sparse primary key codec implementation
support for read, write in memtable
backward compatibility
pluggable codec implementation
codec benchmark
Implementation challenges
No response
The text was updated successfully, but these errors were encountered:
What type of enhancement is this?
Performance
What does the enhancement do?
In the metrics engine scenario, when the number of primary key columns on a physical table becomes excessively large, the CPU overhead for encoding all primary keys increases dramatically. This results in significant performance bottlenecks for both write and query operations.
Proposal
Introduce a sparse primary key codec that encodes only non-null primary key columns. By skipping the encoding of null or empty columns, this change aims to:
TODO
Implementation challenges
No response
The text was updated successfully, but these errors were encountered: