trouble with queryGranularity

hello:

in druid,there are two granularity, AllGranularity, NoneGranularity, corresponding to “ALL”, “NONE”, in the task configuration.

any one who can help me to explan what exact meaning of those.

thanks!

Generally “all” means “put everything in the same time bucket” and “none” means “one bucket per millisecond”. “all” is more useful for querying, you usually don’t want to use it at ingestion time.

hi:

suppose I have some data,such that,

  1. {“ts”: “2015-10-23T12:02:21.234Z”, “k1”:“aaa”, “k2”: 111}

  2. {“ts”: “2015-10-23T12:02:21.234Z”, “k1”:“aaa”, “k2”: 111}

  3. {“ts”: “2015-10-23T12:02:12.001Z”, “k1”:“aaa”, “k2”: 111}

ts: is timestamp

k1, k2 as dimension

if I ingest those data into druid, with task setting:

QueryGranularity.NONE: does it mean log 1 and 2 will rull up as one?

QueryGranularity.ALL: means all three of those log will rull up as one?

Valid values for query granularity also should include :

SECOND,
MINUTE,
FIFTEEN_MINUTE,
THIRTY_MINUTE,
HOUR,
DAY

Hi,

You might want to take a look at this https://github.com/druid-io/druid/pull/1832/files, which added some examples to help explain.