4 ms·
Spark is going to try and ingest all the data, and it won’t fit in RAM. Wrong tool for the job basically.
by mr_toad 8y ago
Spark is going to try and ingest all the data, and it won’t fit in RAM. Wrong tool for the job basically.
- rsynnott 8y agoDepends on exactly how you do it, I suppose, but it shouldn't necessarily. Most Hadoop-y work can also be accomplished in Spark without much fuss.