lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Joel Bernstein (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (SOLR-8963) And new TimeStream to support fine grain time series operations
Date Sat, 09 Apr 2016 18:59:25 GMT

     [ https://issues.apache.org/jira/browse/SOLR-8963?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Joel Bernstein updated SOLR-8963:
---------------------------------
    Description: 
The TimeStream will read Tuples from an underlying stream and expand a unix timestamp into
the individual fields: year, month, day, hour, week, minute, second, milli-second).

This will allow rollups to made on any time grain. This should be very useful for time series
log analysis.

Sample syntax:

{code}
rollup(
            time(search(...,sort="timestamp asc", fl="timestamp,..."), field="timestamp")
            over="year, month,day,hour,minute,second,millis",
            sum(a_i),
            sum(a_f),
            min(a_i),
            min(a_f),
            max(a_i),
           max(a_f),
           avg(a_i),
           avg(a_f),
           count(*))
{code}

Example broken down by customer:

{code}
rollup(
            time(search(...,sort="customer asc, timestamp asc", fl="timestamp,..."), field="timestamp")
            over="customer, year, month,day,hour,minute,second,millis",
            sum(a_i),
            sum(a_f),
            min(a_i),
            min(a_f),
            max(a_i),
           max(a_f),
           avg(a_i),
           avg(a_f),
           count(*))
{code}

To do parallel time series rollups just wrap in a parallel stream and add the partitionKeys
to the search.
{code}
paralllel(..., (rollup(
            time(search(...,sort="customer asc, timestamp asc", fl="timestamp,...", partitionKeys="customer"),
field="timestamp")
            over="customer, year, month,day,hour,minute,second,millis",
            sum(a_i),
            sum(a_f),
            min(a_i),
            min(a_f),
            max(a_i),
           max(a_f),
           avg(a_i),
           avg(a_f),
           count(*)))
{code}












  was:
The TimeStream will read Tuples from an underlying stream and expand a unix timestamp into
the individual fields: year, month, day, hour, week, minute, second, milli-second).

This will allow rollups to made on any time grain. This should be very useful time series
log analysis.

Sample syntax:

{code}
rollup(
            time(search(...,sort="timestamp asc", fl="timestamp,..."), field="timestamp")
            over="year, month,day,hour,minute,second,millis",
            sum(a_i),
            sum(a_f),
            min(a_i),
            min(a_f),
            max(a_i),
           max(a_f),
           avg(a_i),
           avg(a_f),
           count(*))
{code}

Example broken down by customer:

{code}
rollup(
            time(search(...,sort="customer asc, timestamp asc", fl="timestamp,..."), field="timestamp")
            over="customer, year, month,day,hour,minute,second,millis",
            sum(a_i),
            sum(a_f),
            min(a_i),
            min(a_f),
            max(a_i),
           max(a_f),
           avg(a_i),
           avg(a_f),
           count(*))
{code}

To do parallel time series rollups just wrap in a parallel stream and add the partitionKeys
to the search.
{code}
paralllel(..., (rollup(
            time(search(...,sort="customer asc, timestamp asc", fl="timestamp,...", partitionKeys="customer"),
field="timestamp")
            over="customer, year, month,day,hour,minute,second,millis",
            sum(a_i),
            sum(a_f),
            min(a_i),
            min(a_f),
            max(a_i),
           max(a_f),
           avg(a_i),
           avg(a_f),
           count(*)))
{code}













> And new TimeStream to support fine grain time series operations
> ---------------------------------------------------------------
>
>                 Key: SOLR-8963
>                 URL: https://issues.apache.org/jira/browse/SOLR-8963
>             Project: Solr
>          Issue Type: New Feature
>            Reporter: Joel Bernstein
>             Fix For: 6.1
>
>
> The TimeStream will read Tuples from an underlying stream and expand a unix timestamp
into the individual fields: year, month, day, hour, week, minute, second, milli-second).
> This will allow rollups to made on any time grain. This should be very useful for time
series log analysis.
> Sample syntax:
> {code}
> rollup(
>             time(search(...,sort="timestamp asc", fl="timestamp,..."), field="timestamp")
>             over="year, month,day,hour,minute,second,millis",
>             sum(a_i),
>             sum(a_f),
>             min(a_i),
>             min(a_f),
>             max(a_i),
>            max(a_f),
>            avg(a_i),
>            avg(a_f),
>            count(*))
> {code}
> Example broken down by customer:
> {code}
> rollup(
>             time(search(...,sort="customer asc, timestamp asc", fl="timestamp,..."),
field="timestamp")
>             over="customer, year, month,day,hour,minute,second,millis",
>             sum(a_i),
>             sum(a_f),
>             min(a_i),
>             min(a_f),
>             max(a_i),
>            max(a_f),
>            avg(a_i),
>            avg(a_f),
>            count(*))
> {code}
> To do parallel time series rollups just wrap in a parallel stream and add the partitionKeys
to the search.
> {code}
> paralllel(..., (rollup(
>             time(search(...,sort="customer asc, timestamp asc", fl="timestamp,...", partitionKeys="customer"),
field="timestamp")
>             over="customer, year, month,day,hour,minute,second,millis",
>             sum(a_i),
>             sum(a_f),
>             min(a_i),
>             min(a_f),
>             max(a_i),
>            max(a_f),
>            avg(a_i),
>            avg(a_f),
>            count(*)))
> {code}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org


Mime
View raw message