ajantha-bhat edited a comment on issue #3243: [CARBONDATA-3401] fix the java sdk create wrong carbondata filename
URL:
https://github.com/apache/carbondata/pull/3243#issuecomment-544862985
@lamber-ken : In SDK writer, both `taskNo` and `timestamp` are configurable. If user wants milli, user can configure the value.
Also SDK uses nano Second time so that in concurrency the chances that it will be unique than milli seconds.
You might saw that name will have `taskNo, batch_prefix, batchNo, bucketNumber, segmentNo, factUpdateTimestamp`. But in most of the cases `batchNo, bucketNumber` will be zero and `segmentNo` will be null for SDK.
so only differentiating parameters is `taskNo` and `timestamp`. In case of multi-thread SDK writing, `taskNo` also same. same task will write multiple files concurrently. To avoid same name we use nanoSeconds.
In carbon session, this problem will not come as segment will not be null.
So I guess, that current design solves the problem. If we change to milli second, we get issues
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[hidden email]
With regards,
Apache Git Services
Locked
