Hello - I have hard times understanding the benefits of storing (indexing) documents from Spark for a later read into Spark again vs reading using Spark only. How Solr could improve querying time of data through indexing? Is the read operation using Solr faster than using spark.read(someFile.csv)? if yes, is there supporting evidence (e.g., a bar chart comparing the two). Another inquiry is: how could Solr help with querying heterogeneous data through indexing?
- How Spark alone vs Spark-Solr performances compare? marc nicole
