Page tree
Skip to end of metadata
Go to start of metadata

BellaDati IoT Data Collector is a Java based application with micro-services architecture. It is available as a cloud-based service or can be deployed on your own hardware in your infrastructure.

This page describes the supported platforms and system requirements for version 0.6+.

Server

Hardware

In an on-premise environment, a Collector installation consists of multiple distributed modules. These can run on the same computer or on different machines. For evaluation purposes, it is usually sufficient to run all modules on the same computer, e.g. a developer workstation.

Collector hardware requirements mainly depend on the amount of data being processed and the number of concurrent requests during peak minutes. To ensure optimal performance under any circumstances, we group our hardware requirements in 3 tiers depending on usage characteristics. Please note that even so, these configurations can only be guidelines and actual requirements and performance can heavily depend on your usage pattern.

We would appreciate if you could let us know what hardware configuration works in your environment. Please create a ticket in our support system, describing your hardware configuration as well as the number of sensors and data records in your Collector installation.


Minimal Configuration

This configuration is recommended for data throughput up to 100 requests per second (e.g. 100 sensors where each sensor is sending up to 60 requests every minute). We suggest using this for demo or sandbox purposes. Please verify that target machine meets following minimal configuration, because Collector needs sufficient resources (free RAM, CPU, free disk space) to run correctly.

Deployment modeOne machine for all modules

Count of Audit modules

1

Count of Receiver/Sender modules

1-4

Machine

CPU

1x 2 GHz 32/64-bit Intel Xeon or comparable

RAM

2 GB

Disk

1 GB

Network

3G


Medium-load Configuration

This configuration is recommended for data throughput up to 1000 requests per second  (e.g. 1.000 sensors where each sensor is sending up to 60 requests every minute).

Deployment modeOne machine for all modules

Count of Audit modules

1

Count of Receiver/Sender modules

1-10

Machine for Audit, Receivers and Senders (linked and running together)

CPU

4x 2 GHz 64-bit Intel Xeon or comparable

RAM

8 GB

Disk

20 GB, 10,000 rpm, IDE / SCSI

Network

100 Mbps

For production environment, we recommend to build an distributed environment.

High-load Configuration

This configuration is recommended for a up to 1.000.000 requests per second (e.g. 1.000.000 sensors  where each sensor is sending up to 60 requests every minute). It is necessary to run modules on different machines for this configuration.

Deployment modeDistributed

Count of Audit modules

1

Count of Receiver/Sender modules

count of sensors / 1000

Machine for one Audit module

CPU*

32x 2 GHz 64-bit Intel Xeon or comparable

RAM*

60 GB

Disk*

300 GB, 10,000 rpm, SSD

Network*

10 Gbps

*Depends on actual count of sensors and size of one sensor request. Provided value is for 1.000.000 sensors and request size 150 bytes.

Machine for Receivers and Senders (linked and running together)

CPU

2x 2 GHz 64-bit Intel Xeon or comparable

RAM

4 GB

Disk

20 GB, 10,000 rpm, SSD

Network

500 Mbps


Operating System

Collector can run on any operating system for which a supported Java 8+ is available. Such operating systems include:

  • Microsoft Windows Server 2012
  • Microsoft Windows 10
  • Microsoft Windows 8
  • Microsoft Windows 7 (manual installation only)
  • Linux distributions (SuSE Linux, RedHat, CentOS, Ubuntu)
  • Solaris
  • Mac OS X

 

Collector officially supports Oracle Java and IBM Java distributions.

  • No labels