"The page below is from the manual of PRTG, our quick-to-install and easy-to-use network monitoring software"
Try PRTG now and see how it can make your job easier.
PRTG Manual: Detailed System Requirements
This section shows different aspects of system requirements for PRTG. Please consider these requirements to avoid issues while network monitoring.
- Supported Operating Systems
- Hardware Requirements
- Network Size: Recommendations
- Running PRTG on Virtual Machines
- Running PRTG in a Failover Cluster
- Web Browser Requirements
- Screen Resolution
- Requirements for Monitored Devices
- Requirements for the Enterprise Console
- Requirements for Mobile Web GUI
- Requirements for Smartphones
The 32-bit and 64-bit versions of the following operating systems are officially supported for PRTG Core Service and Probe Service :
- Microsoft Windows Server 2012 R2* (recommended)
- Microsoft Windows Server 2012*
- Microsoft Windows 8.1
- Microsoft Windows 8
- Microsoft Windows 7
- Microsoft Windows Server 2008 R2
- Microsoft Windows Server 2008 (not recommended)
- Microsoft Windows Vista (not recommended)
* Windows Server 2012 in Core mode and the Minimal Server Interface are not officially supported.
The version (32-bit or 64-bit) of the PRTG Core Server depends on the version of your operating system. The 64-bit version of the PRTG Core Server will be installed if
- the operating system is a 64-bit Windows system, and
- the system provides 6 GB RAM or more.
Otherwise the 32-bit version of the PRTG Core Server will be installed.
- For best performance of VMware sensors, EXE/Script sensors, and some other sensor types, we recommend Windows Server 2012 R2 on the computer running the PRTG probe: This is either on the local system (on every node, if on a cluster probe), or on the system running a remote probe.
- For best performance of hybrid sensors using Windows Performance Counters and Windows Management Instrumentation (WMI), we recommend Windows 2008 R2 or higher on the computer running the PRTG probe: This is either on the local system (on every node, if on a cluster probe), or on the system running a remote probe.
- Microsoft .NET Framework: We recommend that you provide Microsoft .NET Framework versions 4.0 or 4.5 (with latest updates) on all systems running a PRTG probe. Note: The .NET framework is imperatively needed for monitoring VMware and XenServer virtual environments.
More details: Some sensor types need the Microsoft .NET Framework to be installed on the computer running the PRTG probe. This is either on the local system (on every node, if on a cluster probe), or on the system running a remote probe. Depending on the sensor type, the required versions are 4.0 or 4.5. See the More section for details about the PRTG usage of .NET.
Hardware requirements for PRTG Core Service and Probe Service mainly depend on the sensor types and intervals used. The following values are provided as reference for common usage scenarios of PRTG (based on a default sensor interval of 60 seconds).
A current PC or server with at least a dual core CPU can easily monitor up to 2,000 sensors (see sensor type specific notes below). PRTG supports native x86/x64 architectures.
- RAM memory
Minimum requirement: 2048 MB RAM. You need about 150 KB of RAM per sensor.
- In general, we recommend at least 1 additional CPU core and 1 GB RAM per additional 1,000 sensors.
- Hard Disk Drive
You need about 200 KB of disk space per sensor per day (for sensors with 60 second interval).
- Internet connection
An internet connection is required for license activation (via HTTP or email).
- Stable network connection for remote probes
Remote probes require a stable network connection between the PRTG core server and the remote probe. Unstable connections, for example via 3G, might work but it could be possible that you lose monitoring data if the connection is non-reliable.
There are also non-hardware dependent limitations for some sensor types, for example, WMI and SNMP V3 sensors. These limitations can be overcome by distributing the sensors across multiple remote probes. For clusters we recommend that you stay below 2,500 sensors per cluster.
- Rule of thumb: Typical PRTG installations almost never run into performance issues when they stay under 2,000 sensors, under 30 remote probes, and under 30 user accounts.
- PRTG can scale much higher when the installation is planned well. Please read on if you plan to go beyond these numbers and/or if you plan elevated use of resource intensive features like reporting, xFlow sensors, or clustering.
- If you plan an installation that monitors more than 5,000 sensors from one instance of PRTG on a physical device or more than 2,500 sensors with PRTG running on a virtual machine we ask you to contact our pre-sales team for consultation.
- For your information: To monitor 5,000 sensors in a 1-minute interval, PRTG takes 7.2 million measurements and evaluates, notifies, and stores them—this adds 700 MB of additional data to the database every single day.
- Note: PRTG users usually monitor each device in their network with about 5-10 sensors, so for 50 servers you need about 250-500 sensors.
Apart from the processing power required for the monitoring itself, several aspects can affect the number of sensors that you can use with PRTG. The following recommendations are for a PRTG single core setup (without clustering) on a physical machine. You can overcome some of these limitations if you distribute the sensors across multiple remote probes.
We recommend that you use Windows Server 2012 R2 to run the PRTG core server and probes. This offers superior performance for monitoring, especially if you have more than 2,000 sensors.
We recommend that you mainly use 1-minute scanning intervals for up to 2,000 sensors and 5-minute intervals if you have more sensors.
Number of Users
We recommend that you stay below 30 active user accounts for each PRTG core server. You can work well with more users if these do not all use the UI at the same time (including public dashboards).
Number of Remote Probes
Our general recommendation is to stay below 30 remote probes on one PRTG core server. PRTG still scales well up to 60 probes as long as you have less than 100 sensors per probe.
CPU Intensive Features
Try keeping the usage of the following features down: Many quickly refreshed dashboards, frequent generation of huge sensor reports, heavy usage of packet sniffing, factory sensors, and toplists, frequent automatically scheduled auto-discoveries for large network segments, constant queries of monitoring data via the API, among others.
Network Connection Quality
The quality of your network also plays an important role. When monitoring via UDP, for example, a high packet loss rate can lead to frequent timeouts. Remote probes that connect via unstable (WAN) connections can lead to delays as well.
In general, consider the following rules for the performance impact of different sensor types:
- SNMP V1 and V2, Ping, Port, and HTTP
We recommend that you use these sensor types for scenarios with thousands of sensors.
- SNMP V3
You can monitor about 5,000 SNMP V3 sensors with an interval of 60 seconds on a common two core computer, and about 10,000 sensors on a four core system (the main limiting factor is your CPU power).
Try to keep the number of WMI sensors per probe below 120 sensors (with 60s interval), or 600 sensors (with 300s interval).
- xFlow (IPFIX, NetFlow, sFlow, jFlow)
Monitoring the maximum number of sensors depends on the traffic pattern, the number of xFlow packets per second received by the PRTG probe, as well as the performance of the probe system.
- Packet Sniffer
These sensors create the highest CPU load on the probe system. This technology is only recommended for monitoring of low traffic connections (<50 Mbit/s steady stream). When traffic is often over 10 Mbit/s a dedicated remote probe should be used.
- VMware Monitoring
Monitoring of VMware is limited to about 20 sensors at a 60 seconds monitoring interval, or 100 sensors at a 5 minutes interval. On probes running on Windows Server 2012 R2, you can use more VMware sensors. These limitations issue from the VMware platform. For details please see More section below and refer to the Knowledge Base article "Increasing Maximum Connections for VMware sensors".
- Other sensor types
The impact of a specific sensor type on performance is indicated by a color range when adding a sensor to a device. It ranges from dark green (very low impact) to bold red (very high impact).
To overcome any limitations mentioned above you should distribute the sensors over two remote probes or more.
You can run the PRTG core server as well as PRTG remote probes on virtualized platforms. However, we strongly recommend that you use a dedicated physical machine to run the PRTG core server or the PRTG remote probes. There are several reasons why we recommend that you run PRTG (core server and remote probes) on real hardware, especially for thousands of sensors. Each sensor request will have to go through many virtualization layers, which costs performance and makes measurements less exact. In our experience, a physical machine simply works best for a thousand sensors and more.
Our recommendation to use real hardware is valid for the PRTG core server and for remote probes. If you must run PRTG on a virtual machine, please stay below 2,500 sensors per virtual machine and consider running several PRTG core server instances instead.
Important note: We cannot offer technical support regarding performance and stability problems for PRTG installations on virtual machines that run with more than 5,000 sensors. In this case, please migrate PRTG to one or more, preferably physical, machines.
Note: When running PRTG on a virtual machine, do not use dynamic resource allocation, but please make sure that full resources are available to the virtual machine at any time. In our experience, dynamic resource allocation is not working efficiently for a monitoring software and can therefore lead to massive performance issues.
Please see also the More section for information on running PRTG on VMware.
We recommend a single failover setup if you need fail-safe monitoring. This consists of two PRTG core servers, each working as a cluster node.
In a PRTG failover cluster, the monitoring load doubles with each cluster node, so you will encounter half performance for each additional cluster node. In a single failover cluster, please divide our recommended numbers from above in half.
The following browsers are officially supported by the PRTG web interface (in order of performance and reliability):
- Google Chrome 44 or later (recommended)
- Mozilla Firefox 39 or later
- Microsoft Internet Explorer 11
A screen resolution of at least 1024x768 pixels is sufficient for most functions of PRTG. However, we recommend a screen resolution of 1200x800 or higher.
- SNMP monitoring
The monitored device(s) must be equipped with SNMP Version 1, 2c, or 3 (an SNMP-compatible software must be installed on the device). SNMP must be enabled on the device and the machine running PRTG must be granted access to the SNMP interface. For details, please see section Monitoring via SNMP.
- Windows/WMI monitoring
To use Windows Management Instrumentation (WMI) monitoring, you need a Windows network. For client PCs monitored with WMI, only the operating systems as given above are officially supported. Please do not use Windows Vista or Windows 2008 R1 for WMI monitoring (they both have WMI performance issues). For details, please see section Monitoring via WMI.
- xFlow (IPFIX, NetFlow, sFlow) monitoring
The device must be configured to send NetFlow data packets (NetFlow version 5, 9, or IPFIX) or sFlow packets (version 5) to the machine running the PRTG probe. For details, please see section Monitoring Bandwidth via Flows.
- Packet Sniffer monitoring
Only data packets passing the local machine's network card can be analyzed. Switches with so-called 'monitoring ports' are necessary for network-wide monitoring in switched networks. For details, please see section Monitoring Bandwidth via Packet Sniffing.
- Other sensor types
Depending on the specific sensor type, you can find requirements (for example, modules, components, device configurations) which may have to be fulfilled in the corresponding manual section, as well as when adding the sensor to a device.
The optional PRTG Enterprise Console runs under all supported Windows versions. It has a built-in webkit browser engine and requires no specific browser installed on the system. See also Enterprise Console—Requirements for Connections to PRTG Web Server(s) .
The feature-limited mobile web interface is optimized for low bandwidth and mobile access. It has been designed for and tested with iOS, Android (including BlackBerry), and Windows Phone devices, and also with deprecated Internet Explorer versions. For details, please see section Mobile Web GUI.
PRTG supports optional mobile apps for iOS and Android devices, as well as for Windows Phone. For more information and system requirements, please see section PRTG Apps for Mobile Network Monitoring . The 'Mobile Web GUI' interface can be used on most older smartphones and tablets with a browser that do not support the available PRTG apps for mobile network monitoring. We successfully tested it on Symbian, BlackBerry, Windows, Android, and iOS devices.
Paessler Website: System Requirements for PRTG Network Monitor—Recommended Setup for Most PRTG Users
Knowledge Base: How can I speed up PRTG—especially for large installations?
Knowledge Base: My WMI sensors don't work. What can I do?
Knowledge Base: Frequent Questions about xFlow, Packet Sniffing, Traffic Monitoring and Cisco
Knowledge Base: Increasing Maximum Connections for VMware Sensors
Knowledge Base: My SNMP sensors don't work. What can I do?
Knowledge Base: Checklist for Running PRTG on VMware
Knowledge Base: Which ports does PRTG use on my system?
Knowledge Base: Which .NET version does PRTG require?
Knowledge Base: Why can't I save my PRTG password in Google Chrome?