Open Source Projects:

Case Study

Hazelcast IMDG at British Gas – Internet of Things Case Study

Pricing
Chat
Contact
Back to top

 

For over 200 years, British Gas has provided energy and home services to homes and businesses within the United Kingdom. As one of the largest energy providers in the UK, British Gas also funds energy performance partnerships for businesses and consumer-targeted energy-efficiency initiatives for individual communities.

About Hive

Currently, the UK’s largest Internet of Things (IoT) network, Hive serves over 200,000 homes with a system that allows users to remotely control their heating and hot water temperature from their mobile device or on the Web. Backed by British Gas, Hive hopes to maintain its rapid expansion and use this device as an entry point into the larger sector of complete connected home automation solutions.

Each Hive Active Heating system includes a wireless thermostat, a hub that attaches to an existing broadband router to allow the system to connect to the Internet, and a receiver that connects to the broiler to execute any changes to the desired temperature. The Hive app not only allows users to change their heat settings on the go but also allows users to create a customized heating schedule based on daily routines and holidays. It also automatically prevents frozen pipes and prompts users to adjust their heating needs if it senses the user has left or is returning home.

Challenge

The Challenge

British Gas-funded, start-up Hive needed a technology solution that would allow it to store large amounts of data in memory for quick access. The company also needed a way to integrate that memory store with its current core backend platform and wanted a product that could linearly scale to fit its growing needs. Of primary importance was the ease of querying, the ability to query in real-time, and the capability to initiate custom queries. When evaluating its requirements, Hive also realized it didn’t need persistence or to be concerned with disk I/O. What the company did need was the ability to cache state information for millions of devices, with a simple way to deploy and manage it.

When searching for a solution, this large IoT network company investigated companies such Apache Cassandra and Couchbase. Finally, Hive chose MongoDB to handle their database needs, but after months spent deploying the cloud-based system, the start-up found that it couldn’t meet their platform’s demands: “Although [MongoDB] worked on paper, we were starting to struggle to get the system to cope with the amount of traffic we were generating,” stated Hive. “Since this is a core part of the platform, coming up with a better solution was paramount to us.”

When searching for a solution, this large IoT network company investigated companies such as Apache Cassandra and Couchbase. Finally, Hive chose MongoDB to handle their database needs, but after months spent deploying the cloud-based system, the start-up found that it couldn’t meet their platform’s demands: “Although [MongoDB] worked on paper, we were starting to struggle to get the system to cope with the amount of traffic we were generating,” stated Hive. “Since this is a core part of the platform, coming up with a better solution was paramount to us.”

Solution

The Solution

With one database solution discarded and deadlines looming, Hive decided to try Hazelcast’s in-memory data grid. Hazelcast immediately demonstrated its position as the Web’s leading open source in-memory data grid by performing the fastest in Hive’s initial scale tests.

Unlike other companies, Hazelcast allowed Hive to implement a cache-as-a-service layer. This allowed application caching and eliminated bottlenecks. Soon, Hive was completing 20,000 writes per second in a 20-node cluster with plenty of space capacity and an average latency of under 1 millisecond.

“Moving to Hazelcast proved a big success,” reported Hive, “and we’ve achieved both the desired throughput… and the ease of management and adding capacity. It’s as simple as spinning up a few more nodes and adding them to the cluster — a far cry from the management hell we had with Mongo.” As Hive advances into the remote home automation industry, the company intends to keep using Hazelcast to provide the caching ability integral to their device’s success and use it for general-purpose cache as needed.

Benefits

The Benefits

Hive struggled to find a technology partner that could provide the high throughput and scalability they needed to thrive. Now that the company has found Hazelcast, it’s ready to take on new challenges. Currently, Hive is launching a family of products that tie into Hive’s app and allow users to control lights and appliances remotely. The company also acquired another IoT business, AlertMe, to further cement its presence as a home technology provider.

“[Hazelcast] deploys very simply and just works,” notes Hive. “Nodes cluster automatically and capacity is added linearly with each new node…Our state cache is now a reliable part of the system, with barely any maintenance required.” Pleased with the responsiveness of Hazelcast’s team, Hive has added Hazelcast support to further protect their growing business by providing professional support. The company is also planning to add more nodes to accommodate its rapidly expanding business.

Like many companies, British Gas start-up Hive’s core competencies were not in operational in-memory computing. While Hive had the backing of one of the most successful energy companies in the UK, it still needed to find an expert that could handle its memory store needs. Without the right technological support, the company required several extra months to launch its finished product. Now with Hazelcast’s solutions in place, Hive can focus on delivering a superior home automation product. Hazelcast’s solutions helped create a technological environment that allowed Hive to thrive, and now this company is poised to enter thousands of more homes within the next year.

Hive may have had unique technological challenges, but it had a common problem that many businesses face. Without the right technology to provide an easy-to-use and scalable memory store platform, the company couldn’t focus on the other needs of their growing business. Once Hazelcast provided the solution to their problems, Hive turned its focus back to its core competencies and achieved consistent and encouraging growth. To find out what Hazelcast’s open-source in-memory data grid technology can do for your business, please contact us.

Free Hazelcast Online Training Center

Whether you're interested in learning the basics of in-memory systems, or you're looking for advanced, real-world production examples and best practices, we've got you covered.