Skip to main content


Fig. 13 | Journal of Big Data

Fig. 13

From: Leveraging resource management for efficient performance of Apache Spark

Fig. 13

Processing time using different storage levels of RDDs persistence. This figure shows the results of experimental simulations to assess the RDD persistence by choosing several storage levels. ‘memory_only_ser’ level has less execution time compared to the other levels, because it stores serialized data in memory, it provides more space efficient, and it raises the CPU overhead

Back to article page