Main Article Content
Increasing volume of data that stored in a relational database can cause a reduction in data access speed. The addition of data and database queries can affects the speed of the performance of the information system. The use of document-oriented, non-relational (NoSQL) databases such as MongoDB can help in grouping data logically. Documents and collections do not need to be determined well before making, because it is very flexible. Built-in features of MongoDB have high scalability and are compatible with a variety of software. Data conversion is a very scalable solution, because data will be stored in cloud-based storage, thus saving storage space and server maintenance costs. The latest MongoDB feature also supports multi document ACID transactions to support data integrity. Data conversion case study will use data from the UMKM information system (Usaha Mikro Kecil Menengah) that has been developed previously with a MySQL relational database. The UMKM information system database will support a new database based on NoSQL MongoDB, in order to improve access performance, increase scalability, save on server maintenance costs and help strengthen data integrity. MongoDB database performance test results using the YCSB Client with an increased number of threads affect the linear increase in throughput (ops / sec) and the amount of data.