New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Target Spark 1.5.2 as default Spark version #911

Closed
fnothaft opened this Issue Jan 7, 2016 · 4 comments

Comments

Projects
None yet
3 participants
@fnothaft
Member

fnothaft commented Jan 7, 2016

Or alternatively, Spark 1.6.0. I don't have a strong opinion, but I feel like we've validated more on Spark 1.5.2.

@fnothaft fnothaft added this to the 0.19.0 milestone Jan 7, 2016

@heuermh

This comment has been minimized.

Show comment
Hide comment
@heuermh

heuermh Jan 7, 2016

Member

Want to default to Hadoop 2.6.0 or later at the same time?

Member

heuermh commented Jan 7, 2016

Want to default to Hadoop 2.6.0 or later at the same time?

@fnothaft

This comment has been minimized.

Show comment
Hide comment
@fnothaft

fnothaft Jan 7, 2016

Member

Yes, Hadoop 2.6.0 probably. Good call.

Member

fnothaft commented Jan 7, 2016

Yes, Hadoop 2.6.0 probably. Good call.

@NeillGibson

This comment has been minimized.

Show comment
Hide comment
@NeillGibson

NeillGibson Jan 8, 2016

Contributor

Hi,

Moving to Spark 1.6 would have the benefit that you can migrate the Adam library to use the DataSet API.
But its probably wise to first see how stable the DataSets API is and the uptake of Spark 1.6.

I am not sure if it's worth moving to the DataFrame API in Spark 1.5.2. Since the DataFrame API is incompatible with some transformations /code in this library (e.g. genotypes.toVariantContext). And you would need to support multiple versions of Adam / Spark for the RDD, DataFrame and DataSet API?
#912

Spark 1.5.2 is now widely used but I guess people will move fast to Spark 1.6 if the DataSet API turns out to work / be stable?

Spark version usage
https://databricks.com/blog/2016/01/05/spark-2015-year-in-review.html

Since there is no currently large (production) user base for the Adam library it's easy to move fast to Spark 1.6 / the DataSet API? (and drop support for older Spark versions?)

Contributor

NeillGibson commented Jan 8, 2016

Hi,

Moving to Spark 1.6 would have the benefit that you can migrate the Adam library to use the DataSet API.
But its probably wise to first see how stable the DataSets API is and the uptake of Spark 1.6.

I am not sure if it's worth moving to the DataFrame API in Spark 1.5.2. Since the DataFrame API is incompatible with some transformations /code in this library (e.g. genotypes.toVariantContext). And you would need to support multiple versions of Adam / Spark for the RDD, DataFrame and DataSet API?
#912

Spark 1.5.2 is now widely used but I guess people will move fast to Spark 1.6 if the DataSet API turns out to work / be stable?

Spark version usage
https://databricks.com/blog/2016/01/05/spark-2015-year-in-review.html

Since there is no currently large (production) user base for the Adam library it's easy to move fast to Spark 1.6 / the DataSet API? (and drop support for older Spark versions?)

@heuermh

This comment has been minimized.

Show comment
Hide comment
@heuermh

heuermh Feb 15, 2016

Member

Closed by #932

Member

heuermh commented Feb 15, 2016

Closed by #932

@heuermh heuermh closed this Feb 15, 2016

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment