This post is also available in: Spanish
Towards the end of 2016 we made a short introduction to network monitoring and we told you about the main characteristics to keep in mind when selecting a network monitoring tool. This was meant for users whose installation couldn’t conform with standard syslog monitoring or standard bandwidths.
To see what characteristics we talked about in order for you to make a smart choice, you can refer to that article about network monitoring. In addition, read this article to get more understanding of a network monitor.
- Right time to deploy network monitoring
- Benefits of network monitoring
- Network monitoring checklist (What you should double check)
- Comparison of network monitoring tools
Well, today we’re going to choose the best network monitoring systems available now on the market (both free or licensed) and we’ll compare them. Net monitoring is one of the most important sectors in network tools. What we’ve done on this ocassion has been to use our experience with network monitoring tools, we’ve mounted those that were unknown to us, and we’ve searched other rankings and sources such as:
The objective isn’t to show that Pandora FMS is the best tool in network monitoring, but instead we mean to give an objective vision so you can choose the best tool to adapt to your needs. Every installation is a world apart, and not all network tools are valid for a specific environment.
[UPDATE Dic 2016] Due to the success of the present article, and the number of responses we’ve received, with queries about the different network monitoring tools available and our criteria for choosing the right one for you, we’ve decided to update and add to the information from the first article.
If you’re reading this, it’s most likely because you are thinking of implanting a network monitoring tool or changing the one you already have. [UPDATE Dic 2016]
When is the right time to deploy network monitoring?
To ascertain whether this is the article you’re looking for, we’ll take a look at the conditions that determine if you need to install a network monitoring tool. Obviously, if you are already using network monitoring in your organization and just wish to modernize the tool you use at the moment you’ll know what we’re going to talk about here; just go ahead to the next section (Benefits of network monitoring) if this is you.
For anyone else who is considering the topic for the first time, ask yourself:
- Does my organization depend on network availability?
- Is managing an ever-expanding IT infrastructure becoming overly time-consuming?
- Is the growth of network traffic in my infrastructure exposing bottlenecks? Is the network scaled to optimize all my resources?
Benefits of network monitoring
At the most fundamental level monitoring is about avoiding company downtime, periods of unproductivity, and reducing costs. If you’re not exploiting your resources fully then you’re losing time and money. With the right tool, you can reduce downtime and costs and save money; an interesting concept in the majority of cases. Even if you’re not motivated by profit, any organization (ONG, healthcare, public transport services…) can equally benefit by reducing lost time and streamlining their infrastructure.
Network monitoring and monitoring allow you to optimize both processes and resources; by giving you a global perspective of your infrastructure you can see which nodes are bearing the strain of traffic, where you might need extra hardware and where your system is not scaled correctly. You can see where data is clogging up your system, and why and also what to do to sort out the situation.
Network monitoring provides a heads up on when a problem is coming down the pipe; how to avoid said problem, or stop it from turning into a headache.
Network monitoring tools are also a supplement for your network security, allowing you to detect malicious traffic, where it’s coming from and how to cancel it.
A good monitoring tool also generates time-stamped data logs, allowing the system administrator to build up a collection of historical data that is invaluable both for analyzing how problems have been solved in the past, how your network responded to previously-logged events and allowing for prediction of how your system will behave in the future.
Network monitoring checklist
If you’ve read this far then it’s fair to say you’re in the market for some network monitoring software. But it’s not a straightforward process, and choosing the wrong network monitoring tool now means losing time and money in the future. Furthermore, a sub-optimized tool results in an over-complicated network ecosystem. Here at Pandora FMS HQ, we’ve seen many a costly and complicated installation, the result of a bad choice when implementing monitoring software.
To make sure you’re taking the right points under consideration, here’s a handy checklist:
- Check that the software you’re interested in doesn’t only monitor networks, but that it can also scale up in the case of your network expanding: also, that it can monitor other elements of your organization, including other software such as apps, or hardware like servers, routers or switches: can it give feedback on customer experience, or monitor business processes like transactions, payments, orders? Making the right choice is the difference between growing smoothly, keeping everything under the oversight of a single tool, or incorporating more third-party software and losing the global, unified oversight a good network monitoring tool provides.
- Keep in mind the question of compatibility; are the servers and applications you’re already running going to be compatible with the new network monitoring software?
- Be careful with licenses. Some companies take advantage of clients scaling up to introduce new licensing costs. Enquire about the cost of any additional licenses in the case of introducing new elements or components into your infrastructure.
- Your software configuration and its management are very important. Some monitoring tools involve complex, almost bespoke, configurations that later require a more costly level of competence on behalf of your system administrator.
- Apart from the functional scalability already mentioned, network monitoring can also cover 1000s of nodes and other elements under its aegis. In an interconnected world, businesses need to monitor other devices that are connected to its infrastructure, such as cell phones, vehicles or cash registers. In cases such as these it is vital that your monitoring tool can provide monitoring to all these 1000s of extra devices with maximum efficiency and performance. Ask your provider about their product’s scaling and performance capacities.
- To incorporate all these inevitable extra devices find out if your chosen tool comes with an API capable of integrating with other applications.
- Alerts management. While it isn’t a problem managing alerts while you’re sat at the dashboard of your tool, but what about when you’re not? Your chosen tool should be able to use multiple communication channels (email, SMS, messaging platforms) and that it’s flexible enough to handle both the contents and the requirements of each platform.
- If you’re working with virtual systems it’s imperative that your network monitoring software is also able to monitor those as well. With the idea in mind of expanding your virtual systems, does your tool monitor all your virtual and also the virtual infrastructure itself? Ditto for containers.
- Do you need to inventory the components of your infrastructure? To see them in a simplified way? Do you need to monitor offsite elements? If the answer is “Yes”, keep that in mind when you’re comparing.
- Monitoring embedded systems is another point to consider if you’re monitoring hardware that doesn’t use conventional systems. In Pandora FMS, for example, we’ve monitored fleets of buses, allowing operators to know the status of each vehicle thanks to monitoring their mechanical hardware.
- Another aspect to bear in mind is generating and delivering reports. You need to show the status of the network to non-technical people, in a clear, exportable format, and directed at specific recipients who need different information. A one-size-fits-all is not acceptable here, which is why some monitoring solutions include customizable reports.
- Monitoring with agents or agentless monitoring? The age-old question of whether you need to deploy agents will depend on the nature of your IT environment. Sometimes it isn’t possible to deploy an agent, and sometimes it’s the best solution on offer. Will your chosen monitoring tool allow you to decide?
- Remote monitoring. If the networks to be monitored are geographically, or organizationally displaced, remote monitoring is your friend and ally. Keep disparate networks under observation from a central GUI.
- Cloud monitoring. If all, or part, of your infrastructure isn’t already in the Cloud, it will be sooner or later. Does your network monitoring tool allow for hybrid monitoring (your own CPD and Cloud installation), and to consolidate everything in the same dashboard?
- Saving and analyzing historic data is a crucial element of your network monitoring tool. It’s not only important to know what’s happening in real time but also to analyze past data, in order to make better informed decisions, and to modify your tool accordingly. Network monitoring is based on learning from historic metrics.
- Finally, the GUI. Is it adaptable to your current and future requirements? Will its functionalities scale up as your system does?
Note: all tendencies commented on this article are measured on a worldwide scale.
Steps to follow when evaluating a monitoring tool
In the previous section we looked at the most important characteristics a network monitoring tool should possess. While it’s true that every organization has its own necessities, keep in mind that not all these functions will be applicable or necessary in your particular case. To decide on the most suitable tool foe you:
1. Decide which of the above functions your monitoring tool has to cover
Prepare an Excel with a list of your needs. If your organization has other requirements not mentioned in the above list, don’t forget to put these on the list too.
2. Weigh up the alternatives
At this stage you need to be documenting yourself thoroughly on the available tools; reading, checking out forums, speaking to others in the industry. Concentrate on a shortlist of the aptest tools available, those that adapt to your organization and its economic possibilities, and carry out an in-depth study (using the above checklist as your guideline).
3. Contact suppliers
If a monitoring tool is within your budget, and you have a shortlist of candidates, the next step is to contact the suppliers and outline your project and your needs. Speaking to someone from the commercial department will give you an idea of how used they are to dealing with your particular requirements.
4. Real-world evaluation
If possible, put the tool through its paces in a real life situation, to see how well it adapts to your environment, and if it scales correctly to the dimensions of that environment. This will give invaluable feedback, and provide comparative data if you try out more than one tool. Have your system administrators involved if possible during these trials, as they can provide useful feedback, and a hands-on perspective.
5. Decision time
When it arrives, you will be a lot better prepared to take it for following the steps outlined above.
In depth comparison of popular network monitoring tools
Before going ahead with the review of the main network monitoring tools, we will link to our in depth comparisons related to those network monitoring tools where you can find more detailed information. We will add more reviews as soon as we get time to work on them. If you are interested on a particular comparison, please, let us know writting a comment.
Best network monitoring tools
The open version is capable of monitoring over 10,000 nodes and covers (without limitations) network, server (both with agents or remotely), and application monitoring. With features full of reports, alerts, and third party integrations through API, etc.
Differently from others, it doesn’t have Nagios core as a starting point. Instead, Pandora FMS created its on architecture from scratch, which allows perfect scaling for large environments. A network with over 10,000 nodes has been monitored with Pandora without performance issues (with the Enterprise version though).
We also highlight its integration on mobile devices, which not only allows access to the console, but also to monitoring, thanks to its geolocation system.
Its network autodiscovery system is capable of finding all the elements that compose your network in a short time. One of the key differentiation of Pandora FMS is that it is more than a network monitoring or server monitoring tool. It can monitor whatever you have in mind and it can scale to monitor even business processes showing the global picture of the organisation in the same tool.
You can find more information in http://pandorafms.com/network-monitoring-tools
Probably the best known free tool, and it comes to no surprise since they’ve been working in the U.S. since 1996 to build this monitoring software. Its core is the most important part of the tool and it allows you to build plugins within that core, to monitor particular elements.
It’s interesting to see how the demand tendency on the Internet has been diminishing with the passage of time. What before was one of the most potent and well-known network tools, is losing terrain.
It’s large-scale use is due to the fact that it was the first one to develop a tool that covered unexpendable characteristics in a network monitoring process. For this reason, Nagios was very popular. Furthermore, given its great initial market penetration, it’s still quite used.
- There are a lot of professional profiles with Nagios experience
- If there is a good knowledge of the tool involved, manual configuration can turn Nagios into a very powerful tool to monitor isolated or particular cases
- It has a large plugin offer to adapt the product to the user’s final needs
- Basic configuration is very easy
- Editing or configuration processes are difficult due to the necessity to make manual modifications in order to properly set up the tool
- The GUI lacks user-friendliness.
- Steep and costly learning curve
- Every installation ends up being a “puzzle” where rather than having a standard product, we finish with a personal appliance with hundreds of patches, self-made or third-party codes. All this apart from it being complicated to evolve or maintain by said third-parties.
- Simple reports
- Very lacking when it comes to SNMP, both in polling and trap management.
In a nutshell, Nagios was the origin for monitoring and, as a matter of fact, lots of new network monitoring tools have inherited the Nagios code and made it evolve. Even though you may have a lot of professional profiles on the market, these must have a very vast knowledge of the program, and your installation will depend on them 100%. Future migrations may also be complicated.
Open version: yes
Created by a Lithuanian company in 2005, Zabbix is known for being easy to configure and for having a very powerful GUI. It’s performance starts to decline when a large quantity of nodes is to be monitored. It’s agentless monitoring service stands out above the rest in its category, and experience tells us you can monitor up to 10,000 nodes without performance issues.
- It has quite an active community
- At low levels, it’s still very strong and efficient.
- Even though it has been used for large installations, starting at 1,000 nodes its performance can be diminished
- It’s difficult to create and define alert and report templates. Configurations can require many clicks and steps to be completed
- It doesn’t include real-time reporting
- Difficult to purge when there are errors
- Poor SNMP trap treatment
We’ll show you the interest rate graph for Zabbix:
We’ve got the feeling that many Nagios users are moving over to Zabbix because it’s picked up on Nagios’ glove and it begins to have the visibility that Nagios used to have.
As we mentioned earlier, Zabbix is taking Nagios’ relay and starts to appear on many installations. The problem that we can see here is with its scaling on larger CPD’s. We have to tread carefully if our installation has various elements of the same type (for example databases) because their configurations will be complicated.
An American company that reuses different software elements from Nagios, Icinga or Cacti to create a global solution. It has managed to be among the top ranked network monitoring tools thanks to its mashup of other tools.
Our experience with GroundWork hasn’t been bad, but we’ve seen complications when integrating its different modules. Also, it doesn’t have many plugins developed. For larger environments it falls short. It doesn’t show an extensive history when we monitor a lot of nodes and it doesn’t support platforms such as HP-UX or FreeBSD. If you don’t possess a large CPD or lots of free time to tinker, we recommend checking it out since they have an interesting approach.
Zenoss comes from an American enterprise that has created this software to monitor storage, networks, servers, applications, and virtual servers. Its agentless monitoring is what makes it stand out. It has a “Community” version with very few features and a full-feature Commercial license.
- All terrain regarding platforms. It’s capable of cross-platform monitoring
- It presents a very flexible and tweakable dashboard, which is quite powerful as well
- Great flexibility and potential in the event management offer.
- Depending on the installation’s complexity and on what elements need to be monitored, adapting it can be a rough task
- The DataBase layer can get large and heavy in large environments.
- The panel can be slow on certain installations.
- It only has MySQL and its own database, lacking integration with other DataBase systems.
- Topology maps are one of the weakness of the product.
Zenoss may seem interesting if you don’t really want to invest in monitoring software. If you don’t have that many machines and you don’t plan on investing in support or large features, then you can count on Zenoss as one of your network tools. Be careful though, Zenoss version 5 requires a very potent machine to run properly.
Open version: Yes, but it may just be too limited.
Focused toward small or medium-sized companies. It’s for them that this can arise as a great network tool, and for this reason we’re counting it among the 15 best network monitoring tools.
- Includes web transaction monitoring
- Allows the user to monitor common cloud applications such as Amazon or Rackspace
- Wonderful GUI which is both customizable and dynamic
- Real-time reporting
- Very focused on Linux and Windows
- Has difficulties to add ad-hoc monitoring
- Doesn’t have a free unlicensed version
If you’re a small company, this may be the tool that best suits you. We do recommend you use the fully licensed version, since the free one is kind of basic and can be easily rendered useless unless you’re running a blog or similar webpage. If you’re a small company, this may be the tool you need.
Free version: There really is none. Although, they do offer their monitor.us version, but its features are really basic and recommended for simple websites or blogs.
Icinga comes from part of the Nagios core, over which the GUI was improved. It can be integrated with many databases and its REST API interface is outstanding for integrating other applications. It’s focused on complex networks and monitoring protocols, machine resources and servers.
In 2009 a Nagios fork was created, and since then it’s followed its own path. In its last version (released 2014) they tried to fix performance issues by rewriting the core code. We think that it’s taking the same route as Nagios. It was well recieved back in 2009, but after a great launch, it started to fall back in demand terms. This is yet another monitoring tool that uses the Nagios core.
Open version: Yes
Manage Engine / OPManager
Manage Engine belongs to the Zoho Group, the giant enterprise conglomerate from India, and it’s one of those monitoring softwares to keep in mind. It’s tendency regarding demand is slightly on the rise.
- Easy to install
- Very friendly GUI
- Offers a very wide range of features to cover
- Complex configurations that require a lot of documentation. Very steep learning curve.
- Complicated user experience when navigating through its screens
- Alarm levels limited to basics (warnings and critical status)
- Only available on Linux and Windows
- Lacks inventory and event correlation
- Non-existing features for large environments that are accostumed to work with events, deployment on large architectures, etc.
As we mentioned earlier, it’s one of those network tools that we should bear in mind a lot. What we should warn abour is its really steep learning curve, which may be expensive in resources, money and time in the beginning.
Open version: Non-existent
A multiplatform web monitoring tool (mainly focused on Linux, Unix, HP).
- Graphs on Observium are known for their amount of detail and its careful design. It’s interesting for showing command panels on management levels.
- Easy and useable interface
- Capable of monitoring large installations
- We weren’t able to configure alerts on the free version.
It’s a good tool, but it lacks basic features which in some cases we can recommend complementing with Nagios or Cacti.
Open version: Yes.
One of the monitoring tools that’s more focused on hardware, network traffic, and service monitoring. It’s based on Nagios as well.
It has the capability to monitor multiple platforms. It also monitors cloud systems and virtual environments. We can highlight it’s capacity when it comes to large environments and its scaling on these.
From 2008 to 2010 it had a raise in demand. Ever since then, it’s slow descent has begun. Its market is still Sweden, its country of origin.
- Easy to use
- Very good load balancing system
- Hard to extend features and monitoring processes on our own.
- It won’t allow deployments from the console, so they have to be done manually
Open version: Yes, they have a free version
A tool focused on network and application monitoring. It began developing in 2003, and like many other tools seen here, it started from a version of Nagios. It has a free version available.
The demand tendency for OPSView is quite similar to that of OP5, until 2012 it has a large raise, and from then on its decline has been quick.
- Very rigid monitoring panel
- Limited reports that cannot be exported
- The amount of developed plugins is reduced
In our office we personally like this product quite a lot, even if its demand is declining. Be careful with the tool’s performance.
Open version: Yes
PRTG Network Monitor
Network monitoring tool that stands out for its greatly designed and easy to use interface. It has a great vantage point when it comes to configuring alerts flexibly and because of its report generating capacities. The free version (which is NOT open) is limited to monitoring 100 application types.
PRTG is an application that can only be executed on Windows machines as a part of Microsoft Network Monitoring. Anyway, we highlight that the monitoring is multi-platform and is also able to monitor virtual systems and cloud applications. It can also generate real-time reports.
- Very nice interface with awesome data navigation possibilities.
- You can access monitoring from mobile devices
- Reports can be in PDF/HTML formats
- Very powerful and flexible alert system
- Certain plugins require additional Microsoft user licenses, therefore costs can skyrocket easily
- Limited scalability
- Very rigid when implementing its own checks
- Hard to deploy in environments with a complex connectivity
- Quite inefficient when it comes to server or application monitoring
Open version: Yes, but very reduced
Network monitoring tool that excels for its automatic network and node mapping, without the need to manually activate it. This is added to a very powerful GUI that allows you to easily view your network topology and its status. Solarwinds also allows integrating virtual machines in its monitoring.
Its tendency shows that after a crushing success experienced during 2004 and 2005, a descent happened in which they’re now relatively stagnated.
- Great GUI
- Wonderful community backing it
- Mobile device access
- File and manual configuration for alerts
- The report generation system still has room to improve
- It doesn’t have integration with cloud applications such as Amazon
- You cannot perform all actions from the same panel and you’ll always require to access other tools
- Its licensing forces the user to constantly look at every check performed, since licensing is individually done for these.
A very good option for medium sized companies that can afford the license costs.
Open version: No.
Whatsup Gold is one of the best network monitoring tools when it comes to balancing system loads. Its scaling is horizontal and allows the use of various processes to distribute loads. It’s capable of performing some automatic actions based on events that have occurred.
Just like other tools like SolarWinds or Pandora FMS, it has a system to discover networks and topologies.
By creating proprietary scripts you can add and integrate more applications or services within the monitoring.
We can also connect from mobile devices to access the monitoring panel. Alerts can be set to be sent via email, SMS or through other options.
Information is offered in real time.
- Easy to configure and process network discoveries
- Navigation and console are not that intuitive
- Configuration is split between web configurations and manual configurations on the console
- Limited scalability
- Quite poor for monitoring servers or applications
Open version: No.
A 100% open source network monitoring tool, with no licensing requirements. The business model is to offer the software free and monetize it through consulting services, provided by the OpenNMS group, who administrate the product, through various business lines, including training and support services.
As one can see in the graph, demand for OpenNMS has been on a steady decline for the last decade.
- It’s a flexible and high-capacity network monitoring tool, but one of its chief problems is the steep learning curve associated with it, plus an unintuitive GUI.
- Integrates well with alert systems supports like SMS and email.
- An active community around the tool, and a detailed wiki.
- Can perform similarly to other, licensed products, but requires more time to learn to use and to configure
- Requires a “hands-on” style of configuration, based on script editing and XMLs. In this aspect, it resembles Nagios a lot, in that a manual configuration can lead to highly idiosyncratic installations which can be difficult to decipher.
- Importing third-party MIBs can be problematic.
- The reports it is capable of generating are not polished enough for executive/management eyes, being quite basic visually, and without the possibility for fine-tuning intervals. In fact, it can’t present information in real-time, nor fine-tune intervals to below 24 hours.
- Most importantly, OpenNMS, as its name suggests, is focused on network monitoring, and does not cover the full range of services which a more complete monitoring service can provide, such as server or application monitoring, and much less business processes and user experience. Due to its lack of agents, the information it can retrieve from the system under monitorization is limited to what is retrievable through SNMP, and that is not enough when it comes to monitoring applications or elements of server infrastructure. In some cases (not our own) it has been possible to use Nagios agents in an OpenNMS integration, although we can’t consider that an optimal solution.
Open version: Yes.
Microsoft Network Monitor /Microsoft Message Analyzer
Microsoft Network Monitor was discontinued and is now named Microsoft Message Analyzer. Its main feature is to analyze protocols and to have vision on network traffic. We’ve introduced this tool in this report because we now a lot of people search for it and ask us about it. We must know, though, that it cannot compete feature-wise with any of the formerly mentioned tools.
We hope this comparison of network monitoring tools has been useful. As we understand it, we’ve carefully selected the best market options for network tools. We know that there’s a lot of competition and many options, and it is honestly quite difficult to choose the best option.
Most likely, you’ll find the option you need on this list, most network monitoring tools that we mention are meant for small and medium enterprises. It’s difficult to find tools for large companies that actually cut it for them in performance, and that also support different technologies and protocols that this type of company needs. In this category we can mention ZenOSS and Pandora FMS.
You should also take into account that this article discusses network monitoring tools and, today, it’s more important not only to know the status of networks and applications, but to be able to understan how a business works, from the bit that goes along a network cable, yo sales being done at a specific time by a specific company. This point is related to business activity monitoring, or operational intelligence; but that’s better left for another article.