Why you need network monitoring systems and what they can do for you
A collection of articles that takes you from defining technology needs to purchasing options
Editor's note: In part two of our three-part series on network availability monitoring, we look at the essential network monitoring system features every enterprise needs. Part one covered the evolution of network monitoring tools in the enterprise. Part three will compare the leading vendors in the market.
Network availability monitoring systems are often the first line of defense that network managers rely on when applications go down. In the ensuing blame game that occurs among different teams in the IT organization, the network availability monitoring system can prove essential.
In fact, with the right system in place, the blame game doesn't even have to happen. With good alerting and fault isolation functions, the availability monitoring tool can allow the networking team to adopt a proactive posture toward service problems.
The network availability monitoring system market offers a wide variety of tools, ranging from open source technology that is limited in functionality and scale, to large enterprise products from venerable enterprise IT management vendors like IBM and HP Enterprise. When shopping for a monitoring system, the procurement team must map the needs of their enterprise to the network monitoring system features of the dozens of products currently available.
When assembling a request for proposal (RFP) from vendors, there are several essential things to consider: ease of deployment, usability, compatibility with existing infrastructure, scalability and integration options with the broader IT management tool set.
When looking at how a network availability monitoring system is deployed, the first question to ask is about licensing models. Vendors vary widely in their licensing approaches. Some sell separate licenses for different types of monitoring capabilities, while others charge based on how many devices or how many logical objects or interfaces an enterprise wants to manage. Examine this closely to determine how costly the system is and how complex it will be to maintain.
The next consideration is the actual deployment model. The procurement team should determine if the vendor offers best practice guidelines for configuring its system. Additionally, some monitoring systems are deployed as a single software package, while others require separate installations and separate servers for each component, including the database, the polling engine, the analytics engine and the front-end console. Vendors that offer a deployment with multiple components typically do so to enable maximum scalability and flexibility for their customers, but this approach also adds complexity that not all enterprises want.
Ease of use
A procurement team may assume the existing need for a network availability monitoring system will guarantee the system it chooses will be used productively. This is not always true. Research by Enterprise Management Associates revealed the prevalence of enterprise shelfware -- tools that cost significant time and money to acquire and install – that is never used.
The layout and flow of the management console is one of the most important network monitoring system features for networking staff to evaluate. Do the charts and graphs presented in the system make logical sense to the typical administrator? Are the commands intuitive enough? How much training will the system require? In this context, a proof-of-concept evaluation, where network management personnel actually interact with the system, will help ensure it gets adopted and provides value to the enterprise.
Compatibility with existing network infrastructure
Many organizations operate in a mixed-vendor infrastructure environment. For instance, they may have one or two switch vendors and another that supplies edge routers. Yet another vendor might provide the wireless infrastructure. When evaluating network availability monitoring systems, the networking team needs to determine which vendors and platforms the tool supports.
This evaluation might extend beyond network infrastructure. Many enterprises monitor their servers and storage with these systems, too. When examining which vendors and which devices a management system can monitor, the procurement team must scope out the types of technologies it wants to manage. This includes virtualization technologies and cloud services.
Network availability monitoring systems vary widely in terms of the size of the networks they can manage. While some vendors can monitor hundreds of devices, others can monitor tens of thousands. Furthermore, while some systems can monitor only a single location, others can span an entire enterprise.
It is important to determine the maximum size and level of distribution that a network monitoring system can support -- and, more importantly, how the vendor achieves that scalability. One vendor might monitor a large distributed network with thousands of devices using a single server. Others might deploy multiple servers across the enterprise and tie them together with a front-end console. These variations will influence the usability and complexity of the system.
In all likelihood, this network monitoring system will be part of a larger IT management system. The procurement team should evaluate whether the enterprise will require integration with other management products. For instance, the network engineering team may use the availability monitoring system to monitor and manage the network, while the network operations team might use a higher level service assurance system. Integration between the two will encourage collaboration.
Other potential network monitoring system features include network performance management, log analytics, application performance management, network change and configuration management, reporting capabilities and much more. In fact, many network availability monitoring vendors offer a broad suite of IT management tools with varying degrees of integration among them. For third-party platform integration, however, the procurement team should evaluate the technology partner ecosystem of vendors and the types of certifications and joint support these integrations ship with.
Each of the above network monitoring system features should form the outline of the RFP, underpinning the procurement of an availability monitoring system. There is a large amount of variability in the market, and many vendors will promise more than they can deliver. It is important for the networking team to fully understand requirements for a monitoring system and to map those needs to the capabilities of the products they evaluate. The procurement team should get answers to all of these questions and fully test those answers in a product evaluation before making commitments to any product.
Editor's note: This article was updated in August 2016.
What monitoring functionality does
the system offer?
Modern network availability monitoring systems represent a consolidation of several network monitoring features that network managers pull together from multiple individual point tools. Given this reality, availability monitoring systems offer a range of functions. Procurement teams should determine which of the following core functions they need and how well each monitoring system performs those functions.
- Discovery: Networks are complex and evolve over time. Even a well-documented network will have some unknown devices. Monitoring systems use a variety of methods for automatically discovering devices on the network to bring them under management. Monitoring systems also vary in the scope of device types they can discover. For instance, some systems might be able to discover IPv4 devices, but not IPv6 devices.
- Alarming and reporting: This is the most critical of network monitoring system features in a monitoring system. It is also where many vendors differentiate. Given the web of interdependencies that exist in a network, an alarm on one device can trigger alarms on dozens or hundreds more, leading to alarm fatigue. Products that analyze interdependencies and suppress or prioritize certain alarms can make a network management team more effective.
- Fault isolation and troubleshooting: While alarms notify the networking team of the presence of a problem, fault isolation shows where the problem is. Monitoring tools analyze the data they collect from devices to present this information to the user. The next step is to determine the underlying cause of the issue in that location so the problem can be resolved. Vendors will vary widely in how they perform these functions.
- Inventory and asset management: The discovery functions of a network availability monitoring system can be extended to perform asset management functions. Some IT organizations may elect to use this system to track and organize device support and maintenance tasks.
- New technology support: Enterprise infrastructure is always evolving. The IT organization might adopt software-defined networking or private cloud technologies. They might adopt public cloud services or bring new devices under management as part of an Internet of Things project. As these changes occur, the procurement team should determine whether a network availability monitoring system can help with the overall management of these new technologies.
Learn about how to better monitor applications in your enterprise.
Ensure your network monitoring system is usable.
Track the evolution of network monitoring in the networking profession.
Enhance your enterprise's security with network monitoring tools.