Metry – Telemetry and big-data framework
A Portuguese company that operates on the mobile satellite communications market developed a software framework, designed to aid in building typical big data and telemetry systems, providing a simple programming API to store and retrieve telemetry or monitoring information for different assets over time. This framework – Metry, was originally developed for use in satellite communications and therefore responds to highly demanding search performance requirements.
Metry is a software framework, which stores and retrieves telemetry or monitoring information on a database cluster, which can be composed of several storage nodes. Metry provides a simple programming API to store and retrieve the information. The database, indexes and algorithms used by the Metry framework are optimized to hold very large volumes of information, under heavy insert activity, while providing fast query response times. This framework is designed to allow different types of storage engines, either relational or NoSQL databases. This provides the flexibility to switch to a different product with minimal impact to the source code. The main features of the Metry framework are:
- Java - Metry runs on Java Virtual Machines, a widely used enterprise server execution environment, with the largest available range of open-source products.
- Horizontal Scalability – ability to add multiple nodes to the system, increasing storage and processing power.
- High Availability – no single point of failure.
- Sharding – capability of creating partitions of data, each stored in a different node, to increase performance and responsiveness, based on the location of customers or other grouping characteristics.
- Geographical Redundancy – a reasonably up-to-date version of the database can be available in a separate datacenter, in case of major faults.
- Temporal Database – all data is organized by timestamp, and the system is capable of providing snapshots of information at given instants in time, or across time intervals.
- Spatial Database – the capability to store data with GPS coordinates, and find assets near given positions.
- API – a programming API defining a reusable method of storing and retrieving data; the goal is to be able to integrate new input sources in a very straightforward way, and also to hide database and implementation details.
- Automatic Housekeeping - old data is automatically removed, without impact in the overall performance of the system.
- Exclusive use of royalty-free open-source software - Metry relies exclusively on royalty-free open-source technologies, avoiding any licensing costs for third-party products.
The original use for Metry was satellite communications monitoring and telemetry services. Working with the industry leader in mobile satellite communications, brought highly demanding requirements such as acquiring operational data in real-time, with over 5 million records per hour and at the same time, search and retrieve information with high performance.
Innovations & Advantages
Telemetry can bring significant cost reductions to any business, but its implementation can be technically challenging. With Metry framework, it is possible to build an end-to-end solution to collect vast amounts of data from multiple sensors, store it in a high-availability database, and display information using maps and dashboards.
Metry is based on standard technologies and open-source products, which guarantees an affordable deployment and versatility of new telemetry solutions.
This also means that the customers can start with simple proof of concept installations, with low budget risk, and commit to a full scale project later on.
The Metry framework can scale from simple small business solutions, for internal use, to large enterprises wanting to build and sell value added services to multiple customers.
Current and Potential Domains of Application
Telemetry, Asset Tracking, Remote Monitoring, Fishery, Energy Monitoring, Agriculture, Water Management, Communications, Biotelemetry