Internet-of-Things
Hazelcast Powers Internet-of-things Infrastructure
Hazelcast IMDG is used as the back-end infrastructure for many Internet-of-things (IoT) deployments. Everything from in-home thermostats and lighting systems to industrial sensors such as smart-meters and in-store m-commerce systems.
IoT architectures involve massive numbers of data events often coming from millions of devices. Such fast moving data requires in-memory data caching techniques to keep up with the flow in real-time. Moreover, systems must be scalable and elastic in order to handle surges in usage and a rapidly expanding user base. Hazelcast IMDG can ingest writes at very high-velocity and can process the business logic or complex computation on incoming data in real-time using in-situ processing.
Hazelcast IMDG provides an ideal NoSQL solution to meet demanding service-level requirements of IoT architectures, while at the same time providing great simplicity in configuration, deployment and management making it is easy for customers to build and maintain a next generation IoT system.
Why Most Data Stores Fall Short… |
Why Hazelcast IMDG is Ideal… |
| Relational databases cannot keep up with the numbers of events. | Hazelcast can ingest huge numbers events into memory, process them and then update a database behind either in summary or with batches, thus achieving processing but also persistence. |
| For real-time analytics on large volumes of data, NoSQL and RDBMS solutions suffer from latency issues. | As a fully in-memory data store, Hazelcast IMDG can transform and ingest data in microseconds. |
| Many NoSQL solutions such as Cassandra lack the speed required to keep up with today’s transaction volumes, as they rely on disk-based storage. | Hazelcast is an order of magnitude faster than NoSQL including Cassandra. |
| Many databases and data stores cannot be scaled elastically at runtime. | Hazelcast can be scaled elastically by adding new servers without production interruption. In the cloud this can be done in seconds and can even be automated. |
| Most IoT systems are 24×7 and cannot be shut down for maintenance. Almost all relational databases and NoSQL require scheduled maintenance for upgrades, patches etc. | Hazelcast has rolling upgrades, where maintenance activities can be done server by server and client by client without the need to shut the cluster down. |
Why Hazelcast IMDG is Ideal… |
| Hazelcast can ingest huge numbers events into memory, process them and then update a database behind either in summary or with batches, thus achieving processing but also persistence. |
| As a fully in-memory data store, Hazelcast IMDG can transform and ingest data in microseconds. |
| Hazelcast is an order of magnitude faster than NoSQL including Cassandra. |
| Hazelcast can be scaled elastically by adding new servers without production interruption. In the cloud this can be done in seconds and can even be automated. |
| Hazelcast has rolling upgrades, where maintenance activities can be done server by server and client by client without the need to shut the cluster down. |
Customer Success Story
British Gas currently operates the UK’s largest IoT network. It’s Hive Network 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. Hive completes 20,000 writes per second in a 20-node Hazelcast IMDG cluster with plenty of spare capacity and an average latency of under 1 millisecond.
“Moving to Hazelcast IMDG 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 MongoDB.”
British Gas is rapidly expanding its IoT network with the addition of additional heating systems as well as expanding into the larger sector of complete connected home automation solutions.
