[
https://issues.apache.org/jira/browse/ASTERIXDB-1073?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Till Westmann updated ASTERIXDB-1073:
-------------------------------------
Component/s: Optimizer
AsterixDB
> Use an index for comparing datetime to current-datetime()
> ---------------------------------------------------------
>
> Key: ASTERIXDB-1073
> URL: https://issues.apache.org/jira/browse/ASTERIXDB-1073
> Project: Apache AsterixDB
> Issue Type: Improvement
> Components: AsterixDB, Optimizer
> Reporter: Steven Jacobs
> Assignee: Steven Jacobs
>
> Using the following DDL:
> drop dataverse emergencyTest if exists;
> create dataverse emergencyTest;
> use dataverse emergencyTest;
> create type CHPReport as {
> "id":int,
> "timestamp":datetime
> }
> create dataset CHPReports(CHPReport)
> primary key timestamp;
> This query will use an index search:
> for $emergency in dataset CHPReports
> where $emergency.timestamp >= datetime("2012-08-07T10:10:00.000Z")
> return $emergency.message;
> Whereas this will do a data scan:
> for $emergency in dataset CHPReports
> where $emergency.timestamp >= current-datetime()
> return $emergency.message;
> Both are build on BTREEs so both should be optimized to an index search. It
> seems like there should be a way to use the index when comparing to
> current-datetime, Although I suspect the solution is nontrivial because of
> the way that current-datetime is currently working.
> Here are the plans, respectively:
> distribute result [%0->$$5]
> -- DISTRIBUTE_RESULT |PARTITIONED|
> exchange
> -- ONE_TO_ONE_EXCHANGE |PARTITIONED|
> project ([$$5])
> -- STREAM_PROJECT |PARTITIONED|
> assign [$$5] <- [function-call: asterix:field-access-by-name,
> Args:[%0->$$0, AString: {message}]]
> -- ASSIGN |PARTITIONED|
> project ([$$0])
> -- STREAM_PROJECT |PARTITIONED|
> exchange
> -- ONE_TO_ONE_EXCHANGE |PARTITIONED|
> unnest-map [$$6, $$0] <- function-call: asterix:index-search,
> Args:[AString: {CHPReports}, AInt32: {0}, AString: {emergencyTest}, AString:
> {CHPReports}, ABoolean: {false}, ABoolean: {false}, ABoolean: {false},
> AInt32: {1}, %0->$$8, AInt32: {0}, TRUE, TRUE, FALSE]
> -- BTREE_SEARCH |PARTITIONED|
> exchange
> -- ONE_TO_ONE_EXCHANGE |PARTITIONED|
> assign [$$8] <- [ADateTime: { 2012-08-07T10:10:00.000Z }]
> -- ASSIGN |PARTITIONED|
> empty-tuple-source
> -- EMPTY_TUPLE_SOURCE |PARTITIONED|
> distribute result [%0->$$5]
> -- DISTRIBUTE_RESULT |PARTITIONED|
> exchange
> -- ONE_TO_ONE_EXCHANGE |PARTITIONED|
> project ([$$5])
> -- STREAM_PROJECT |PARTITIONED|
> assign [$$5] <- [function-call: asterix:field-access-by-name,
> Args:[%0->$$0, AString: {message}]]
> -- ASSIGN |PARTITIONED|
> project ([$$0])
> -- STREAM_PROJECT |PARTITIONED|
> select (function-call: algebricks:ge, Args:[%0->$$6, function-call:
> asterix:current-datetime, Args:[]])
> -- STREAM_SELECT |PARTITIONED|
> exchange
> -- ONE_TO_ONE_EXCHANGE |PARTITIONED|
> data-scan []<-[$$6, $$0] <- emergencyTest:CHPReports
> -- DATASOURCE_SCAN |PARTITIONED|
> exchange
> -- ONE_TO_ONE_EXCHANGE |PARTITIONED|
> empty-tuple-source
> -- EMPTY_TUPLE_SOURCE |PARTITIONED|
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
