Objenious: Faster Data Processing for the Internet of Things

Objenious is a startup subsidiary of Bouygues Telecom. It offers the first network (LoRaWAN) in France dedicated to connected objects with 84% of the territory covered and 96% of the population.

Its IoT platform dedicated to connected objects, SPOT, for Smart Portal of Things, allows its customers to manage all of their fleet of devices and their data.

A subsidiary of Bouygues Telecom, Objenious offers its services using a LoRaWAN network. With over 4000 antennas installed across the country, LoRaWAN is a low-power wide area network that connects sensors to the internet.

After its launch just two years ago, Objenious very quickly designed its infrastructure. Its predecessor system struggled to manage the amount of data generated by the network, but failed to facilitate the kind of agile development Objenious hoped to achieve.

En interne, nous utilisons des méthodologies Agile qui s’appuient sur un processus DevOps, reposant quant à lui sur l’infrastructure sous-jacente.

En tant que startup, nous devons proposer très rapidement des fonctionnalités à nos clients, c’est pourquoi nous avons eu besoin d’une grande agilité sur le plan technologique

Souchiam Sechao

Directeur technique chez Objenious

Eliminate data transmission delays

The use of smart sensors and IoT technology generates huge amounts of data which is constantly updated and must be stored immediately. As its data volume began to increase, Objenious’s previous infrastructure found itself unable to keep up with demand, and its portal began to experience latency. Objenious needed an infrastructure that was reliable and able to provide its customers with the data insights they needed immediately.

For this, Objenious switched to Google Cloud Platform. Thanks to Google Kubernetes Engine with Docker, Objenious has developed a containerized infrastructure capable of easily adapting to demand. By setting up BigQuery to work with Objenious’s Cassandra database, it becomes possible to automate the transposition of data from one to another. Objenious is now able to offer immediate data analysis to its customers, and analyze its own data.

A "zero downtime" migration

During the migration to the new platform, Objenious’s services had to remain operational. "We had to find solutions in order to maintain the service during the migration", explains Souchiam. Objenious began collaborating with Claranet, who offered invaluable expertise while working with Kubernetes. Souchiam explains: “Our teams already knew each other. So there was a natural affinity between them and Google experts. "

The services were migrated one by one in just 3 months without any disruption. Claranet now offers ongoing support, in collaboration with Google, so Objenious does not need to be involved in handling every request.

As soon as a problem arises, the managers of Claranet and Google can communicate directly, without Objenious present. This was not possible with our previous partners. We offer a much faster and more integrated level of service.

Overall, Claranet has enabled us to achieve our goals of agility and high availability of the platform. This success is obviously due to the proximity of the Claranet teams but also to their expertise and skills around Google Cloud Platform.

Souchiam Sechao

Technical Director at Objenious

Better integration with GCP

The migration to Google Cloud Platform has helped Objenious improve its performance, security and profitability. "We are now able to handle thousands of messages per second without affecting the platform, and adapt our infrastructure according to our needs," explains Souchiam. “And thanks to the optimizations we've put in place, the total cost of ownership is very profitable. We are able to project ourselves into the future on this basis."

Next year, Objenious plans to start using machine learning to improve its data analysis services, and expand its open source projects. The company will also explore how LoRaWAN can be integrated with Google Cloud Platform, and more specifically Cloud IoT Core. “Our better integration with Google Cloud Platform will accelerate the development of all customers who use Google for the management and development of their IoT applications."

Find out more about our Google Cloud Platform outsourcing offers