What is Network Analysis?
Network analysis is the analyzation of networks through graph theory. The network can actually be a traditional computer network, home network, wireless network, transportation network, or biological network, among others. It often acts as a network management tool for breaking down projects into components, or individual activities, and recording the results on a flow chart or network diagram. These results generally reveal information that is used to determine duration, resource limitations, and cost estimates associated with the project.
Network analysis can offer insight into what is occurring at each critical point of the network. Details pertaining to protocols, traffic flows, and individual data packets can help an organization keep its network operating at an optimal level. When equipped with this information, IT teams can detect remote application and network degradations before they become problems. This helps to manage bandwidth more efficiently, improve availability, and minimize the time needed to resolve issues through rapid detection and precise isolation.
The challenge many IT organizations face is gaining access to the essential information that is required to perform network analysis, and to respond accordingly to the feedback extracted. In most cases, it is the lack of IT presence at remote sites that hinders the execution of thorough analysis. Due to the increasing complexities involved with network analysis, organizations may find that, in order to be effective, they require a solution that offers three important qualities: accessibility, granularity, and visibility.
Accessibility is a quality that can be provided by robust network analysis hardware and software, which can be deployed throughout the organization, but is typically maintained from a remote site. For most enterprises, the cost, time, and impact associated with sending personnel to deal with technical issues is both prohibitive and unacceptable.
Granularity refers to the project’s vision into activities and events across all layers of the network. This means that the solution must include far more than traffic analysis provided by standard analytics. It must also include the capture and decoding capabilities needed for network packet analysis.
As it relates to network analysis, visibility is achieved by a solution that automatically identifies and discovers network updates. This better assures that precise analytics can be maintained without requiring IT staff to be dispatched to remote locations. Visibility can be enhanced through the ability to view comprehensive graphs, charts, reports, and network diagrams. Such extensive insight can give an organization the opportunity to quickly identify, and resolve, problems that exist anywhere within the network.
@MrMoody - I did some GIS network analysis for a company once. GIS allows you to conduct geographic statistical analysis and superimpose the results on a map.
For example, you can ask the question, “How many hospitals in such and such a region have more than 1,000 patients?” Instantly the answer will be revealed as a series of shaded regions on the map, dotting the locations that answer the query.
You can do multilevel analysis as well, where you have several layers of data superimposed on each other, all on one map. Executives love this kind of information display.
@everetra - I majored in computer science and well remember an interesting story about network analysis and graph theory. It was called the traveling salesman problem.
In essence, you have this salesman who has to get to a number of different cities in the shortest time possible. Of course, he can take different routes to get to those cities. Which routes are the most efficient for him?
This problem became the basis for a lot of computer algorithms that emerged later, and which attempted to solve problems similar to it.
When you think about it, it has a lot of applications. An airline needs to find the best routes to reach a destination. An overnight shipment carrier needs to find the best routes to ship a package to you on time. Yes, it also includes telecommunications networks and things like that too.
Graph theory can certainly be an interesting study in and of itself.
I worked in the telecommunications industry for over ten years. While I wasn’t an engineer myself, I worked around engineers, and they were constantly engaged in network traffic analysis.
They were monitoring where the bulk of the phone calls went, whether there was enough infrastructure (T1s, for example) to handle the incoming traffic, how the traffic was being routed, etc.
These were mission critical judgments because we told our customers that we would guarantee a certain quality of service for their phone services.
It was all about eliminating or at least reducing congestion on the network so that phone calls would reach their destination as quickly as possible.
Post your comments