cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Pavel Yaskevich (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-11990) Address rows rather than partitions in SASI
Date Tue, 09 Aug 2016 01:12:20 GMT


Pavel Yaskevich commented on CASSANDRA-11990:

I looked at the code again and I think we might be better off keeping only Murmur3Partitioner
(tokens as longs) for now instead of changing so much just to support RP...
That would also make update path much easier for us - a. added clustering positions, b. extend
tokens sizes to variable size. instead of having intermediate of multi-size fixed tokens...

*Code remarks:*

   what is the point of it it's still wrapped into logger.isDebugEnabled() ?
- Can you move drop-data and rebuild related changes to the separate branch so we can keep
changes to token tree and bug fixes separate since they are separate tickets?

   I'm not sure what is this `// ignore` about?...
- SasiToken -> SASIToken since SASI is an abbreviation already.
- Back to the point about not supporting PR; the way TokenTreeSerializationHelper is currently
done is not optimal since it has to be explicitly carried around,
   which means that it's most likely not a proper abstraction to have, e.g. higher levels
like QueryPlan/Operation/Expression/ColumnIndex should never care about any details
   about TokenTree, especially about how it's tokens are serialized, we should either use
IPartitioner interface or nothing at all.
- Upgrade test is missing for TokenTree.

> Address rows rather than partitions in SASI
> -------------------------------------------
>                 Key: CASSANDRA-11990
>                 URL:
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: CQL, sasi
>            Reporter: Alex Petrov
>            Assignee: Alex Petrov
>         Attachments: perf.pdf, size_comparison.png
> Currently, the lookup in SASI index would return the key position of the partition. After
the partition lookup, the rows are iterated and the operators are applied in order to filter
out ones that do not match.
> bq. TokenTree which accepts variable size keys (such would enable different partitioners,
collections support, primary key indexing etc.), 

This message was sent by Atlassian JIRA

View raw message