Will Spark overtake Hadoop? Spark is very popular and much faster then MapReduce. So, can it replace Hadoop?
Hadoop is Big Data platform where Spark is just in-memory processing engine. Spark does not have own data storage system.
Hadoop is used as distributed file system and computation framework.
While Spark is used for in-memory processing of data.
Wherever there is need of MapRecude like program Hadoop will be used. For in-memory processing Spark is used.
So, there no such comparison.
Developer will use both Hadoop and Spark .
Check more tutorials at: