banner

A general purpose platform for IoT and Big Data

HyperIoT is an Intuitive, cloud-ready tool for exchanging, displaying, combining and analyzing large amounts of data from generic sources as well as IoT devices.

  • General Purpose
  • IoT & Big Data
  • Cloud-Ready
  • Intuitive

See HyperIot In Action

HyperIoT provides, even to less experienced users, a robust IoT and Big Data server infrastructure. It offers the ability to navigate data in real time through intuitive graphical widgets, create customized statistics through a wizard

Big Data Services
70%
Custom Statistics
50%
Real Time widget dashboard
80%
PC con Dashboard

How Does HyperIot Works

Speed Layer
Speed Layer

Similar to the Batch layer, it creates views based on the received data, but it "looks" only at the most recent ones. It updates the real-time view every time it receives new data.

Serving Layer
Serving Layer

It saves the results of the batch views calculated by the underlying layer. It's typically characterized by a NoSQL distributed database

Batch Layer
Batch Layer

The lowest layer of lambda architecture. Its goal is to save the raw data and "pre-calculating" the views from these

ARCHITECTURE STACKS

Explore HYPERIOT

HyperIot is agnostic to data sources. It has been chosen to support the MQTT communication format which is the most widespread in the field of IoT. Security mechanisms have been supported in order to reduce the risk of intercepted packets being directly visible.

The section relating to the streaming of the platform is the one that will directly manage the manipulation of the incoming data in real time. The features to implement are different:

● channeling on HDFS (Hadoop Distributed File System)
● manipulation (data unpacking and rewriting with a different scheme)
● serialization / deserialization of incoming data
● enrichment of the data structure

For various reasons, therefore, a different database paradigm was chosen, which is the one given by the so-called NoSQL databases, which renounce the relational nature to allow problems related to large amounts of data to be solved. In particular, a structure was chosen in which nothing is shared between the different nodes of the cluster (shared-nothing), in order to allow optimal scaling. In particular, the relational structure is renounced by exposing a data without a pattern, in which all the information is “exploded” in the table.
Currently when we talk about web applications we mean one of the following: Single-Page Applications, Isomorphic / Universal Web Applications, Hybrid Web Applications, Native Web Applications and recently Progressive Web Applications.
img

Projects in Different Domain of Industry

Flame Sensor

Smart Safety

client-img
client-img
client-img
client-img
client-img
client-img
client-img
Translate »