hadoop - 50-60 gb of data in spark standalone mode -
i trying analyze around 50-60 gb of data. thought of using spark that, not have access multiple nodes in cluster. can level of processing done using spark standalone mode ? if yes, know estimated time required process data.thanks!
short answer: yes.
spark partition file in many smaller chunks. in case few chunks executed @ time. these few chunks should fit in memory (you need play configurations right)
to summarize, able it, faster if had more memory/cores can processes more things in parallel.
Comments
Post a Comment