[jira] [Updated] (CASSANDRA-11990) Address rows rather than partitions in SASI

["Alex Petrov (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/CASSANDRA-11990?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Alex Petrov updated CASSANDRA-11990:
------------------------------------
    Attachment: perf.pdf

Attached is the (very naïve) performance benchmark, although in some cases even this one is indicative. 

Compared are trunk and the [early patch for this branch|https://github.com/ifesdjeen/cassandra/tree/11182-trunk]. 

I'm writing 10, 100 and 1000 rows with 10, 100 and 1000 byte string values. 

On clustering column, the difference is insignificant, since SASI can already take an advantage of clustering restrictions. However, on "regular" column the difference is more obvious.

Querying on static columns currently would result into the whole partition read, even if there's another {{Operation}} intersecting in the current partition. 

> Address rows rather than partitions in SASI
> -------------------------------------------
>
>                 Key: CASSANDRA-11990
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-11990
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: CQL
>            Reporter: Alex Petrov
>            Assignee: Alex Petrov
>         Attachments: perf.pdf
>
>
> 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
(v6.3.4#6332)