Thesis

  1. Home
  2. thesis
  3. mqtt
  4. Thesis description

CLOUDBACKEND – EDGE CLOUD COMPUTING DB PLATFORM AS A SERVICE

layers CloudBackend provide customers with a database in the cloud to build apps on ranging from mobile, web, to IoT. The database is decentralized all the way from cloud and edge to the local device to maximize performance and minimize latency. This enables CloudBackend customers to build solutions for connectivity, mesh networks, edge computing, data management, AI/ML, federated AI, autonomous vehicles, and drones. Customers benefit from a secure, decentralized, synced data model across a low latency distributed network. The CloudBackend SDK transports data to/from any device or app across the cloud, edge, or self organizing network (mesh). The service and technology are a foundational and complimentary technology to get the most out of 3G/4G/5G/Satellites, and AI/ML.

MQTT IOT INTERFACE FOR AN EDGE DBPAAS — THESIS DESCRIPTION

MQTT is a standard for communication across a wide array of IoT devices. It is designed as a lightweight publish/subscribe messaging transport for data. Adding MQTT support to the CloudBackend decentralized database service would mean that existing devices or program code can start utilizing CloudBackend as its database without changing a single line of code. The thesis is all about converting MQTT messages to/from its native protocol format into the capabilities and data models supported in CloudBackend with objects, key/value, hierarchical graphs, and relational data. Decentralized CloudBackend databases can then act as hubs receiving and sending information to MQTT enabled applications, while data is coordinated and synced from the decentralized locations using the greater CloudBackend infrastructure across the edge and cloud to aggregate data.

With this thesis suitable for one or two people, we wish to address the following questions and establish a methodology and implementation for how data resources in CloudBackend can be created, updated, and accessed using MQTT, please see mqtt.org.

  • Figuring out an ideal data model to be used in CloudBackend for MQTT messages for new solutions being developed
  • Provide a methodology how existing data models in CloudBackend can be exposed and accessed using MQTT (possible framework with policies to define MQTT broker behavior)
  • Aim at making the MQTT generic so it can be used without changing existing MQTT implementations or CloudBackend implementations

The thesis should implement or conclude:

  • Implement a MQTT broker that provides a MQTT interface to CloudBackend
  • Ensure existing MQTT applications can be switched to CloudBackend without adoptions
  • Allow CloudBackend to receive MQTT in a format that makes it easy to aggregate the data and apply AI/ML solutions on any decentralized database along the way to the public cloud
  • If required to make the solution work, provide a framework for controlling the MQTT broker using centrally defined policies in CloudBackend (defined as XML documents)

The implementation is required to be in C++ on Ubuntu Linux and should be built on top of the CloudBackend C++ SDK. All work will be done from the CloudBackend AB main office and/or from home in Linköping. Proof of citizenship in one of the European Union states is required.

Contact
Daniel Arthursson, CEO
thesis@cloudbackend.com

∧