Let’s be honest: Network management is a big responsibility.
It includes monitoring and controlling a computer network to ensure all of its resources—both hardware and software—are in good shape and are being used productively. And if you’re not a network expert (or don’t have one on your payroll), that probably sounds pretty intimidating.
But there’s no time like the present to tackle the topic of network management head on. And if you’re looking for a crash course, then you’ve come to the right place. In this article, we cover:
- What’s a network?
- What’s network management?
Let’s get started with a quick refresher on the basics of networking.
What’s a network?
A network is a collection of connected devices—like computers, mobile devices, servers, routers, switches, bridges, firewalls, access points and more—that can communicate with one another through the transfer of data.
To get data from point A to point B, each device in a network plays an important role. Let’s take a quick look at some of the more common network devices and what they do.
Here are some important devices that make up an average network:
- Clients: Not to be mistaken with your MSP’s clients, a client is an application or whole system that connects to a server. In a client-server network model, a client sends a request—think sending an email—to a server, and once the server fulfills a client’s request, the connection is terminated.
- Servers: A server is a device designed to accept and respond to requests made by another computer over the internet or a local network. A server helps manage network resources like sending email, managing print jobs, or hosting a website.
- Switches: A switch is a physical or virtual device that connects all of the devices on a network to each other. A switch enables network devices to talk to each other by sending incoming data from input ports to output ports that will take it to its intended destination on the network—it determines where to send each incoming message by looking at the physical device address, otherwise called the MAC address.
- Routers: A router is a physical or virtual device that connects computer networks. A router analyzes a data packet’s destination IP address, determines the best way to reach that destination, and then forwards it there. It sounds similar to a switch, but Cisco has a way to remember the difference: “Switches create a network. Routers connect networks.”
- Firewalls: A firewall is, by definition, a router—its primary function is still routing, but with additional security rules. These rules help a firewall permit or block data packets coming and going from the network.
- Access points: An access point is a wireless network device that connects endpoint devices to a local area network. An access point can extend the wireless coverage of an existing network, which increases the number of users that can connect to it and changes the way users are able to connect to the network.
A “computer network” can refer to a lot of different things—it might mean a few devices in a single room, or it might mean hundreds of thousands of devices connected around the world.
Those devices can be connected physically, virtually, or wirelessly, while others may be connected by some combination of the three.
For simplicity’s sake, a network is defined based on its purpose or its size. As an MSP, the types of networks you deal with vary from client to client. Here are six common network types your clients likely have:
- LAN: A Local Area Network connects computers and devices together across short distances, like between an office or a few close company buildings, to share resources. They’re one of the simplest—and the most common—networks, and you’ll often be expected to monitor and manage them.
- WLAN: A Wireless Local Area Network works like a regular LAN, but it uses wireless network technology like Wi-Fi so devices don’t need to be physically connected to the network with cables.
- VLAN: A Virtual Local Area Network allows you to use a single physical infrastructure while logically segmenting the network into multiple VLANs. VLANs are used to segment the network, apply security controls, and manage traffic.
- WAN: A Wide Area Network connects computers and LANs together across longer physical distances so they can communicate over one network. The internet is the most common example of a WAN.
- VPN: A Virtual Private Network is an encrypted connection that allows users to send and receive data as if their devices were connected to a private network, even when they’re not. Your clients may give employees a VPN so they can access the company network remotely.
First, let’s make one thing clear: Network topologies aren’t the same as network types. Network topology refers to the layout of a network—it’s the network’s virtual shape or structure and shows how different network devices are connected to each other.
No matter which network topology or topologies exist in your client networks, it’s really important to understand them—and have a detailed, up-to-date document that shows how everything is connected.
Without this information, even basic troubleshooting can be a pain. For more information on network topologies, when to use them, and how to document them, download our free ebook The No Sweat Guide to Network Topology by network engineer Kevin Dooley.
How data gets from one device to another
Connecting the right devices together is only part of the larger networking story, however. Once they’re connected, one important question remains: How do network devices actually communicate with each other?
The OSI (Open Systems Interconnection) model breaks device to-device communication into seven layers:
|Layer 1||Physical||The physical layer connects network devices to one another physically and wirelessly, using things like cables and equipment. It’s important that the right devices are connected to each other with the right cables across an acceptable distance. In the event of a network issue, this is the first layer you’re likely to check.|
|Layer 2||Data Link||The data link layer is where switches operate in most networks. Network devices that are directly connected to each other can transfer packets to one another at this layer.|
|Layer 3||Network||The network layer is where routers operate to efficiently push packets towards the correct destination. For example, if I’m in Auvik’s Waterloo office and I send an email to an MSP in Barcelona, there are millions of paths for my packets to go down. Routers at the Layer 3 level ensure my packets will end up at the right device in Spain.|
|Layer 4||Transport||The transport layer deals with data transfer between end systems and hosts. It determines how much data needs to be sent, the rate it needs to be sent at, where it’s going, and how traffic is segmented for delivery before being put back together.|
|Layer 5||Session||The session layer allows users on different devices to establish “sessions” between each other. A session is created when two devices “speak” to each other, and this layer establishes, maintains, and terminates connections.|
|Layer 6||Presentation||The presentation layer translates the network format to the application format (or vice versa). This layer is responsible for data representation, data encryption, and data decryption, and it ensures systems with different data representation can still communicate with each other.|
|Layer 7||Application||The application layer is what most users see—it’s the web browsers, Microsoft Offices, and social networking platforms of the world.|
The OSI model can help an MSP narrow problems down when things go wrong by being able to work through the layers—is a cable broken, is a router malfunctioning, or is an application experiencing an outage that’s out of your control?
An alternative to the OSI model is the TCP/IP (Transmission Control Protocol/Internet Protocol) model, which is a set of protocols used to connect network devices on the internet (or within a private network). The TCP/IP model uses four layers instead of seven, and it defines how applications can create communication channels across a network.
Now that we’ve covered some of the key networking basics, it’s time to examine how it’s related to network management and your job as an MSP.
What is network management?
Network management is the process of monitoring and controlling a computer network to ensure all of its resources—both hardware and software—are in good shape and are being used productively.
This process is overseen by a person—a network administrator, network manager, or MSP—who uses protocols, systems, devices, and applications to ensure the network can keep up with business demands and user expectations.
The challenge in network management is that networks are becoming more and more complex. There are more devices on networks to manage than ever before (thanks in part to a massive IoT boom), and they’re not all created equal. Even similar devices on the network—say a switch and a switch—can be drastically different if they’re manufactured by different vendors.
Network management is a huge responsibility. To better understand what the job entails, it’s helpful to break things down into five key areas encompassed by the FCAPS acronym: fault management, configuration management, administration, performance management, and security management.
The “Big 5” network management functions
- Fault management
- Detect the problem.
- Isolate the problem and determine what’s causing it.
- Troubleshoot and resolve the problem.
- Document the process that was used to resolve the problem.
- Configuration management
A fault—more commonly called an issue or an error—is an event that shows a problem in the network. Common faults are things like packet discards, packet errors, or high interface utilization, but the possibilities are essentially endless.
The goal of fault management is to minimize downtime. So the faster a network manager knows about an error in the network, the better.
To fix a fault before it escalates and interferes with client uptime, there are four important steps a network manager needs to follow:
For MSPs, step 4 is extremely important. With multiple techs troubleshooting issues, documenting previous faults and how they were resolved can help streamline resolution in the future. This way, your team isn’t wasting time trying to solve the same faults repeatedly with no reference point.
For a network and its devices to do what they’re supposed to, they need to be configured or told what to do.
The goal of configuration management is to monitor and document network and device configurations. This requires a network managers to set, maintain, organize, and update configuration information for both the network and its devices.
Network configurations include:
- Router and switch configuration , which specifies the network’s correct IP addresses, route settings, and more.
- Host configuration, which connects a host device (like a computer or laptop) to the network by logging network settings like IP address, device routing tables, and DNS server configuration.
- Software configuration, which gives appropriate credentials to network-based software—like Auvik, for example—so it can access the network and monitor network traffic.
Network device configurations include operating system version, firmware version, serial numbers, IP address, routing configuration, switch VLAN configuration, and more.
To keep track of all this information, it’s best practice for network managers to securely store this information. Then, as configuration parameters change, you want the documentation to immediately update so the information you have is always up to date.
This documentation is a critical part of configuration management. It’s referred to when a network or one of its devices needs to be modified, expanded, or repaired, which helps return the network or device to its pre-outage state quickly and minimize downtime.
The “A” in FCAPS can mean administration or accounting.
As an MSP, you’re responsible for network administration. This means administering network users with passwords and permissions, backing up software, and more.
Accounting management refers to resource usage and billing—this is important if you’re an MSP who offers internet services, bills on the usage of a WAN link, or monitors public cloud usage.
Performance management is all about the efficiency of the network—both right now and in the future.
Network performance is measured using metrics like throughput, uptime and downtime, error rates, percentage utilization, response time, and latency. This data is primarily collected using SNMP (Simple Network Management Protocol), without which we couldn’t identify devices, monitor network performance, or keep track of changes to the network.
To effectively monitor network performance, a network manager:
- Gathers performance data—manually or automatically.
- Analyzes the data to establish baseline levels.
- Establishes performance thresholds. This step feeds into fault management—when a threshold is exceeded, it means there’s a problem that needs to be taken care of.
Sometimes, data may seem abnormal. To get extra visibility and context into performance metrics, you can dive beyond the performance metadata SNMP provides to get closer to the traffic by leveraging protocols like NetFlow.
By collecting and analyzing NetFlow data on a client device, you’re able to get important information from incoming and outgoing traffic. This can include information like source and destination IP, source and destination port, and Layer 4 protocol. With this information, you can determine if network traffic is legitimate.
This data is important for forecasting the future of the network, otherwise called network capacity planning. If the client is seeing internal bottlenecks while only doing business-critical work, its likely they’ll have to upgrade their network infrastructure. If they’re using all of their external bandwidth, then they’ll need to increase the capacity provided by their ISP.
The data can also be used to help make the case for network upgrades on the fly. If you discover a network device (or devices) consistently causing performance issues, then it’s likely a sign it needs to be updated or replaced.
Today, there’s no shortage of cybersecurity threats—from DDoS attacks and cryptojacking to mobile malware, it feels like we’re all one click away from taking down whatever network we’re on. That’s why security is a critical part of network management.
The goal of security management—and the main responsibility of a network manager—is to prevent, detect, and quickly respond to prevent security threats. Since 100% prevention isn’t possible, a quick response to security threats is essential.
This is achieved through using different security tools and systems, including:
- Firewalls: A firewall is a device that monitors network traffic and uses a set of security rules to permit or block data packets coming and going from the network.
- Reverse proxy servers: A reverse proxy server masquerades as a web server on the edge of a network, while the real server sits somewhere inside the network. It forwards data to and from the real server, which is only needed if you have a server accessible from the internet.
- Web application firewalls: A web application firewall is a device that sits between the internet and a web application to protect applications from cross-site forgery, cross-site-scripting, SQL injection, and more.
- Intrusion detection systems: An intrusion detection system monitors packets and sessions for malicious activity. If activity is discovered, it’s reported.
- Intrusion prevention systems: An intrusion prevention system detects and prevents malicious incidents. It reports any threats to a system admin or a network manager and takes preventative action, like closing access to the network.
- Email scanners: An email scanner protects endpoints from spam and malware attacks by scanning emails for viruses.
- Antivirus software: Antivirus software prevents, detects, and removes software viruses from network devices.
These tools and systems help to authenticate, encrypt, and authorize traffic flowing through the network and ensure it’s supposed to be there.
If traffic is unauthorized, policies can be applied to prevent users from gaining access to sensitive information. And if unauthorized traffic makes its way through, a network manager has a much bigger issue on their hands.
As an MSP, security management can also extend to physical security, as electronic access and surveillance systems are connected to the network.
What can I use to help me manage a network?
It’s impossible to effectively carry out the Big Five network management functions without systems, tools, applications, and processes to help.
A network management system includes tools and applications that support a network manager in monitoring and controlling the network. A network management system can detect, configure, monitor, and troubleshoot network devices, mitigating the need for a lot of manual work.
While there are specialized applications that can help you carry out each function, you’ll ideally have one that helps you cover all five. Yet not all network management systems are created equal.
So, with so many tools on the market, how do you know which one is right for you? Here are five things you should ask yourself before investing in a new network management system.
Can it automate manual processes?
To keep pace with your constantly evolving client networks, it’s important to leverage automation wherever you can. Luckily, there are a lot of manual network management tasks that can be automated, including:
- Network inventory and assessments
- Network mapping and device discovery
- Configuration backups
- Wire tracing
- Remote network infrastructure access
If a network management system can automate these traditionally time-consuming tasks—and ideally more—then you’ll be able to focus on other things, like winning new clients and keeping them connected.
Does it give real-time visibility into the network?
For many network management systems today, real-time visibility is no longer a negotiable feature—it’s a must-have.
You need visibility to identify issues before they escalate and cause downtime, and to quickly troubleshoot issues when they do interrupt productivity. With real-time visibility, network managers can immediately see the operational health and status of the network and its devices, and be alerted when an issue arises.
Does it offer multi-vendor support?
It’s your job to keep all your clients’ network devices and workstations connected. For someone unfamiliar with network management and managed services, it might not seem like that tall of a task.
But consider this: For network devices alone, there are hundreds of different vendors, and any given client likely has a mishmash of devices in their network. In fact, a majority of MSPs are managing 4 or more different network device vendors across their clients.
Throw in a lack of industry standardization across devices and things get pretty complicated.
In order to manage mixed-bag environments, a network management system needs to be vendor-agnostic, or offer multi-vendor support.
Can it integrate with other tools?
After identifying which automated, vendor-agnostic tools will best help you achieve your network management goals, the next step is to look at how—or if—they’ll work together.
There are tons of tools that fit each FCAPS component (quick refresher: fault management, configuration management, administration, performance management, and security management) category.
If you can’t find one that covers all five, then you’ll want to integrate the ones you use into the rest of your software stack as much as possible.
Pro tip: Look for products and vendors that have opened up their tools with APIs (application programming interfaces). APIs give you the flexibility you need to integrate other products into your software stack.
With an integrated software stack, you’re able to make your workflow more productive and efficient—and make managing a network a heck of a lot easier.
Ready for a more in-depth look at network management? Start by checking out these five helpful resources:
- The Art of Monitoring by James Turnbull
- Computer Networking for Beginners: Your Guide for Mastering Computer Networking, Cisco IOS and the OSI Model by Ramon Nastase
- Network Maintenance and Troubleshooting Guide: Field Tested Solutions for Everyday Problems by Neal Allen
- Network Management: A Complete Guide by Gerardus Blokdyk
- Network Warrior by Gary Donahue