Effective network monitoring can be a key to success for your business. The past few years have seen network administrators deploying networks, obtaining the facilities to run them, and instituting network security that covered internal operations. With much of that deployment now over, today’s focus has turned to more sophisticated network functions like network monitoring.
Companies are often driven to improve their network monitoring by the regulatory guidelines that govern their industries. This is especially true if you are in financial services or healthcare. This article discusses network monitoring, and some of the options that are available to small, medium and large-sized organizations.
The three basic areas of network monitoring are:
Asset Management: Tracking and accounting for all network assets throughout their life cycles, from installation to removal;
Operations Management: Tracking of network activities, device readiness status, all current versions of software releases, areas/devices where software/hardware updates are needed, and confirmation that all devices on the network are authorized;
Change Control: Tracking, recording and auditing of all changes administered to the network and its devices.
“In the past, a LAN was just another thing to keep up and running, but the issues today are much more complex than just connectivity and access,” says Dave Boulos, vice president of Product Management at ComBrio, a supplier of secure connectivity middleware to third-party hardware manufacturers and service providers. “In the asset management area, the LAN administrator must control all of the devices on the network. If devices are added or removed, where did they get added to or removed from?”
The practice of operations management, as well, is more than pinging a device. You need to know the version of software that your network is running, and the state of all devices. Strong network operations management also ensures that there are no any rogue devices on your network. It will tell you where upgrades are needed, and whether licenses are up to date.
Change control is another major IT initiative. Network monitoring tools allow you to take snapshots of the network and can even be automated to take network snapshots whenever there is a change. These changes become an audit trail for purposes of tracking and accountability. They allow a LAN manager to troubleshoot problems that might have occurred because of a specific change.
Making it Work
It’s easy to recognize the benefits of network monitoring from an IT standpoint. What is more problematic is developing the justification for network monitoring tools and capabilities required for the corporate budgeting process.
“For many companies, the cost of directly purchasing network monitoring tools is a barrier to entry,” says Boulos. “The IT manager must be able to go to the business manager with a return on investment, and this is hard to do when the initial investment for a large enterprise can be as high as half a million dollars.”
Boulos says that for large enterprises, there are also network decisions that must be known in advance, such as whether a VPN will be used.
Companies considering their network monitoring options can look at several different business models:
Turnkey network monitoring: You purchase and install your own internal monitoring solution for the network that you administer;
“Pay as you go” network monitoring: You contract on an as-needed basis with an outside service provider who supplies the network monitoring services to you for a charge;
Fixed network monitoring: You contract on a regular basis with an outside service provider to supply you with network monitoring services;
Partial network monitoring: You use a combination of internal administration of network monitoring and contracting out certain network monitoring functions to an outside service provider.
Outsourcing network monitoring provides advantages because you do not have to expense or capitalize a costly network monitoring solution or worry about training staff to work with it. Instead, you contract out the services and pay a fraction of the cost to an outside provider.
However, like any outsourcing strategy, assigning network monitoring to a third party means that someone on your staff must be well-trained in the area of vendor management. This person must be able to hold the vendor to specific objectives and SLAs (service level agreements), meeting regularly with the vendor, ensuring the financial and operational health of the vendor, and also the security of the information that the vendor will hold or be aware of.
If your company is in a highly regulated industry, all of these vendor issues are sure to be on the examiner’s checklist when he pays you a visit.
Monitoring the Enterprise
The choices you make for your network monitoring depend on your core business, the industry you are in and the size of your organization.
For example, if your core business is online retail and the Internet is your sales channel, network monitoring should be on the top of your list because business revenues depend on the network being up all of the time. You also have significant customer information and privacy issues. Companies functioning in this area often opt to purchase and operate their own network monitoring systems because the risk of not doing so is too high, and it is easier to cost-justify a sizable investment in network monitoring solutions at budget time since the end business is highly dependent on reliable networking.
A second issue for companies is the industry that they are in. If the industry is highly regulated, companies are going to be held accountable for a strong network monitoring platform. This platform can be attained through either internal investment and administration of network monitoring–or by contracting with a reputable service provider with a track record of reliability, accountability and financial strength.
The last tier to consider is the nature of the network itself.
“In an SME, a LAN administrator has a closed infrastructure,” says Boulos. “He knows everyone who is on the LAN is authorized to be there, and he understands all of the network elements that are behind the firewall. Conversely, a large enterprise has people who are not only in headquarters, but also in branch locations and home offices.
“An enterprise might have to lease lines to locations for purposes of security, or to set up VPNs between locations. For the large enterprise network administrator, the responsibilities of network monitoring are multiplied. The best approach in that case is a centralized monitoring system.”
What to Watch For
Although companies are at different places, most have noted the network monitoring issues that are hot in today’s environment:
* corporate security;
* evolving network standards;
* the need for IT to stay up to speed with evolutions in both the security and the standards areas;
* ease of network management;
* the importance of network change control and asset management.
“In the end, LAN administrators must make sure that the LAN is accessible to the business because if it isn’t, it can affect the business and the revenue stream,” Boulos says.
An effective network monitoring solution must give the network administrator all the tools he needs to perform and produce the necessary reports and audit trails of network activity.
Mary E. Shacklett is president of Transworld Data, a marketing and technology practice for technology companies and organizations.
Network monitoring factors to watch
If you listen closely, your network is trying to tell you something. Monitoring your network–or the backbone of most any operation–can be relatively simple and should ideally be a vital component of any business, including e-commerce Web sites, telecommuter-driven companies and home-based businesses with an online presence.
On a basic level, monitoring your network’s availability allows you to notice if your Web site or key hyperlinks are down well before your customers, prospects or stakeholders notice. When you monitor availability, you take ownership of your network and make a conscious decision to not rely solely on your vendor to ensure your network operates at optimum levels.
Here are some key practices to keep in mind:
* Track speed: Monitor how quickly–or slowly–the pages of your Web site load. A cursory and informal check of your site may reveal that your site does, in fact, load quickly. But what about when you’re not actually on your site? How does it load at 7 a.m., 8 p.m. and on the weekends?
Your Web server needs to constantly run at optimum levels for your site to truly load quickly at all times of the day. By constantly monitoring your network with an automated application, you can identify the true average speed of your site and points in the day when marked fluctuations or slowdowns occur.
Beyond your site’s pure download speed, your site may be slow to load, because there are too many images–or large images–on your Web site. Using a Web site-analysis tool to pinpoint large image files on Web pages is an essential component of optimizing your Web site.
* Track usage: Monitor how much of your allocated bandwidth you’re using overall and for specific functions such as accessing the Internet, transferring proprietary data and storing files. There’s no way to accurately determine your exact bandwidth needs without knowing precisely how much of your current bandwidth is and isn’t being used.
You may uncover that your service provider is inadvertently overcharging you for sporadic fluctuations in your network traffic or sudden dramatic increases in network traffic.
Perhaps most notably, tracking network usage allows you to make informed and strategic purchasing decisions. If, for instance, you observe a constant rise in your network usage, it may be more prudent to upgrade your network before you hire more staff or add a bandwidth-draining product gallery to your site.
* Set network alarms: For network monitoring to be truly effective, a company ought to set up instant-notification mechanisms that streamline the network-alarm process and send an e-mail, sms, or instant message to a systems-support employee or contractor who can help correct the network failure or problem.
* Have a Plan B: In a perfect world, mission-critical systems such as your e-mail servers wouldn’t go down. But, in all likelihood, they will at some point, so you ought to have an umbrella solution in place that will essentially allow you to re-direct your traffic to a standby server.
For instance, you may consider running a nightly updated copy of your main Web site on a second, dedicated Web server located at a different location from your primary Web server. If there’s a problem of any kind with your primary server that adversely affects your Web site, you can simply change your site’s DNS entry to move all traffic to the backup server.
* Test your plan: Prevention and theoretical solutions to mission-critical problems aren’t enough to safeguard a company from an actual network failure or outage. Testing is critical. If, for instance, you plan to move customer traffic to a backup Web server during a network failure, then you should test whether the backup server can, in fact, handle the traffic and allow your site to operate at acceptable levels. — Dirk Paessler, CEO, Paessler AG