Performance Comparison of CRUD Methods using NET Object Relational Mappers: A Case Study

Abstract

Most applications available nowadays are using an Object Relational Mapper (ORM) to access and save data. The additional layer that is being wrapped over the database induces a performance impact in detrimental of raw SQL queries; on the other side, the advantages of using ORMs by focusing on domain level through application development represent a premise for easier development and simpler code maintenance. In this context, this paper makes a performance comparison between three of the most used ORM technologies from the .NET family: Entity Framework Core 2.2, nHibernate 5.2.3 and Dapper 1.50.5. The main objective of the paper is to make a comparative analysis of the impact that a specific ORM has on application performance when realizing database requests. In order to perform the analysis, a specific testing architecture was designed to ensure the consistency of tests. Performance evaluation for time responses and memory usage for each technology was done using the same CRUD (Create Read Update Delete) operations on the database. The results obtained proved that the decision to use one of another is dependent of the most used type of operation. A comprehensive discussion based on results analysis is done in order to support a decision for choosing a specific ORM by the software engineers in the process of software design and development.

Nesne-İlişkisel Eşleme (ORM) Araçlarının .NET 6 Ortamında Performans Analizi

“…In this scenario NHibernate performs better in terms of processing time with all record count groups including 500, 1,000, 2,000, 5,000, and 10,000. On the contrary to our insert operation results, the results of [4] may show that for complex scenarios, NHibernate may have better performances. Similarly with processing times, memory usage statistics in [4] are not coherent with our study.…”

“…The test environments, databases that are used and operations are not standard within these studies. For example, in [4], one of the insert scenarios includes three insert operations to the tables in one-to-one relationship. In this scenario NHibernate performs better in terms of processing time with all record count groups including 500, 1,000, 2,000, 5,000, and 10,000.…”

“…al. [4] compare Dapper, EF Core and NHibernate tools. These studies shed the light on the performances of different ORM tools.…”

“…Zmaranda et. al [4], measures the RAM usage besides processing times. A study by Balcı [5], which is not performance comparison research on ORM tools but examines the performance analysis of the Entity Framework ORM tool on different databases, included all of the processing time, RAM and CPU usage information.…”

Case Study of Self-Organization Processes in Information System Caching Components

“…Modern Information Systems use caching mechanisms, as the process of their design almost always use the technology for mapping relational data into object model [1]. This technology allows leveling down semantic gaps during system development, but also creates additional expenses in processor resource and memory consumption during the operational phase [2,3,4,5].…”