Khanan: Performance comparison and programming α-miner algorithm in column-oriented and relational database query languages

Abstract

Process-Aware Information Systems (PAIS) support business processes and generate large amounts of event logs from the execution of business processes. An event log is represented as a tuple of CaseID, Timestamp, Activity and Actor. Process Mining is a new and emerging field that aims at analyzing the event logs to discover, enhance and improve business processes and check conformance between run time and design time business processes. The large volume of event logs generated are stored in the databases. Relational databases perform well for a certain class of applications. However, there is a certain class of applications for which relational databases are not able to scale well. To address the challenges of scalability, NoSQL database systems emerged. Discovering a process model (workflow) from event logs is one of the most challenging and important Process Mining tasks. The α-miner algorithm is one of the first and most widely used Process Discovery techniques. Our objective is to investigate which of the databases (Relational or NoSQL) performs better for a Process Discovery application under Process Mining. We implement the α-miner algorithm on relational (row-oriented) and NoSQL (column-oriented) databases in database query languages so that our application is tightly coupled to the database. We conduct a performance benchmarking and comparison of the α-miner algorithm on row-oriented database and NoSQL column-oriented database.We present the comparison on various aspects like time taken to load large datasets, disk usage, stepwise execution time and compression technique.