JISA (Jurnal Informatika dan Sains) (Jan 2020)
Optimasi Jalur Transfer Data dari HTTP menjadi MQTT pada IoT menggunakan Cloud Services
Abstract
At this time, in the era of technological development and information that incresingly very fast, introduced smart device systems that can help any stuff and conditions of human work, this technology is known as IoT or the abbreviation of Internet of Things Technology. With the existence of IoT systems, every devices connected to microcontroller, and we can communicate to domonitoring stuff and work more easier anytime, anywhere, and with variety of gadgets that we have. However, it must be supported by the existence of a stable supporting internet access. The HTTP protocol is used to connect communication networks and data traffic between devices on IoT has laxity with effectiveness, supports faster data transmission capabilities, resources intensive used, complicated message formats, large data sizes, and the process of checking traffic for each data consume a lot of time. This exactly affects for the performance of related systems. Referring to these constraints, this paper will discuss the comparison betw en HTTP and MQTT protocols in terms of protocol models, speed of data transmission, size of messages sent, and optimal use for each protocol. Implementation of this system test also involves the implementation of RabbitMQ as a Message Broker between the sender / publisher (smart phone) and the receiver / subscriber (IoT device) which is tasked with storing temporary data before being distributed to the receiver. Not only that, RabbitMQ also has a role to handle if the internet connection is unstable or temporarily dead, the server can store data temporarily on RabbitMQ and will return the data to the IoT device if the internet connection is more stable. This such a system mechanism can certainly be a solution for us to against the problem of data loss due to unstable internet connections. Users/ clients are no longer afraid that data will be lost because all data will be accommodated in RabbitMQ stored place, not influenced by the stability of the internet network connection. In this case, RabbitMQ must be in container environment, for that we also need a docker when installing a broker which is useful as a media containerize. With the docker container, installing RabbitMQ becomes so easy because the installation is not really done from the beginning (as we knew that potentially quite complicated), but we will only install containers that have been filled by RabbitMQ.
Keywords
