GoTraxx In-Memory Stream Analytics Platform

Traditional platforms that use the “batch” methodology require that the data be completely available and stored in a database before processing can begin. With GoTraxx’s “In-Memory stream analytics”, the data is made available as soon as it arrives at the system, before being stored in a database, making it possible to manage the data ultra-fast in “real-time”.

Use cases

Your business is made up of a continuous series of events, vehicles move, generate alarms, make accidents, deliveries are completed or fail, vehicles break down and enter the workshop.

Your Fleet management program requires “real-time” data management and alarms that only “stream processing” can offer. You cannot base your choices on what happened yesterday or the past month. You need to react quickly to what’s happening now. Proactive management depends on modern “in-memory stream analytics” platforms for:

• Analyze data in real time
• Alarms
• Fraud management
• Incident management
• Delivery management
• Management of technical downtime
• Visualization of the fleet in real time

Continuous-flow processing must be both fast and scalable to allow millions of records to be processed every second. Each data flow is potentially an infinite sequence of events that must be correlated with external data sources and then processed in real time.

The steps that make up the GoTraxx platform data processing include:

• Receipt of data and cleaning of the same
• Enrichment through external databases (Reverse Geocoding and Meteorology)
• Time correlation with previous data
• Analysis and generation of alarms and statistics
• Publication of the data via MQTT and Websocket
• Database storage for historical analysis

In which fields of my business do I need real-time data?

GoTraxx’s data flow processing capabilities are based on a distributed “in-memory computing” technology which, leveraging the access speed of the RAM memory compared to that of a disk, is placed between the source of the data and the destination of the itself understood as an App, Website or External Systems, and is offered in cloud or “on-premises” mode.

Our customers use the capabilities of our platform to capture and process data in a few milliseconds to identify anomalies, respond to alarms or simply know where their vehicles are at that exact moment.

The management of the data in a “stream” way does not always eliminate the need for a traditional type of processing (“batch”). “Batch” processing remains necessary when you need historical data such as reports, analysis of distances or the driving behaviour of a driver.

It is for this reason that aggregated data with the same atomicity that is required by the user are stored in a database, but only and only the data necessary for the business, making access to this data source as fast as possible.

Given that it may be necessary to analyse historical data with algorithms other than those used at the time of their processing, the GoTraxx system stores the source data received from the different sources in a “Data Lake” so that nothing is lost.

The components of the GoTraxx platform

The IOT Mediator

Data comes from different sources and with different methods, GoTraxx is capable of receiving data via TCP, UDP, MQTT, REST, Kafka and new adapters can be developed to increase the entry points. The IOT Mediator takes care of receiving and translating them from its own language into a language common to the entire platform.

“In stream” data processing engine

The data processing engine is composed of a set of nodes that process the data simultaneously and in a distributed manner, coordinated with each other by a coordination system that allows horizontal scalability and redundancy at the same time.

Distributed storage

The GoTraxx platform uses different types of storage in addition to RAM to allow the processing of events in milliseconds and ensure data redundancy.

Time Frame

GoTraxx processes the data flow in terms of “time windows”, where a “time window” represents a set of data coming from the same device over a specific period of time. This time window is always stored in RAM and this makes possible the processing speed that the platform is capable of.

Time stamp of the event

GoTraxx temporally marks each event directly where it is generated, in the event that the original order of events is not respected during processing through different nodes, a component checks the digital signature of each packet and temporarily parks the subsequent pending events laggards, restoring the original order of the data.

Fault tolerance

GoTraxx manages fault tolerance through memory “snapshots” for each node and each component, this allows the process to restart automatically and, using the “snapshot”, restart exactly where it left off.

A high-speed solution for a high-speed world

In-memory streaming is designed for today’s digital ecosystem, with millions of source streams it is the only one to guarantee a service without ever a delay.

The speed of the RAM memory optimized for the management of IoT data flows

The speed guaranteed by “in-memory” systems can accelerate data processing performance by a factor of 1000X. Adding to this the “stream processing” speeds up even more, making the data available before it is stored in a database

Applicable to every area of ​​your business

Is there even a single area of ​​your profession that would work better at a slower speed? The GoTraxx platform is not only faster, it is enormously faster, opening the door to new opportunities for innovation.

If you want to access all the benefits of GoTraxx, fill out the form below to activate your free 30-day trial now.

---