Why dynamic mapping is changing network troubleshooting for the better
- 16 May, 2017 07:10
Effective network troubleshooting requires experience and a detailed understanding of a network’s design. And while many great network engineers possess both qualities, they still face the daunting challenge of manual data collection and analysis.
The storage and backup industries have long been automated, yet, for the most part, automation has alluded the network, forcing engineering teams to troubleshoot and map networks manually. Estimates from a NetBrain poll indicate that network engineers spend 80% of their troubleshooting time collecting data and only 20% analyzing it. With the cost of downtime only getting more expensive, an opportunity to significantly reduce the time spent collecting data is critical.
Visibility into the network can drastically simplify how network engineers collect data when they troubleshoot. Currently, most network engineers use the command-line interface (CLI) as the window into the network.
The CLI only allows engineers to troubleshoot as quickly as they can issue and interpret commands, working through each device one at a time. CLI is a high-powered spotlight into specific areas of the network, but what network engineers really need is to light up the entire room.
Most organizations work through CLI and static diagrams created using MS Visio. Using the CLI in conjunction with an accurate, up-to-date network diagram provides needed visibility into the network, but doesn’t provide insight into the details that can truly make a difference. In the event of an outage, engineers need to understand live performance and recent network changes – not just topology and configuration.
Without effective visual aids that can detail networks in real-time, engineers are left guessing and checking each device in the network.
Dynamic maps and real-time visibility
What troubleshooters really need is a customized diagram that omits irrelevant network sections so they can effectively hone in on the source of the issue. If a slow application is affecting three data centers, an engineer needs a single diagram of the application flow, not three diagrams, one for each data center.
Most organizations are still troubleshooting and leveraging static diagrams while automated, dynamic mapping can easily simplify the process. Organizations should be leveraging dynamic maps that can show network performance in real time. Dynamic maps not only allow engineers to enter two IP addresses (the source IP and the IP of the application server), but also include an analysis of routing, access-lists and Network Address Translation (NAT) for every step of the map.
This saves hours in diagnosing and data collecting time as engineers can easily see the source of the issue and address it directly.
Attempting to troubleshoot performance problems with no visibility is challenging. Many organizations opt for 24x7 network monitoring systems, which generate alarms when an incident occurs. These solutions identify when there is a problem, but don’t offer visibility into the specific area of the problem.
Organizations that integrate automated dynamic mapping solutions with a network monitoring system are quickly alerted when there is an issue and can identify the source of the issue much faster than organizations that fall back on manual networking.
Change can be risky
The simplest way for any network engineer to identify the source of an issue is to understand what’s changed about the network. Without dynamic mapping, it’s a difficult, if not impossible, task.
A NetBrain poll showed that more than one third of network outages result from a change in the network, and visibility into the specific change relieves much of the troubleshooting pain for engineers. This allows engineers to see bandwidth bottlenecks or CPU/Memory overutilization in real time without the painstaking process of manually diagramming a network and manually testing each device.
Automated network diagrams also allow for real-time comparison. Network engineers can quickly compare the existing network diagram to a network diagram from before the incident, giving specific data on the potential problem.
Automated network diagrams create a single source of truth for network engineers, providing a clear windowpane into an increasingly complex world.
The demands on networks are only increasing with the adoption of software-defined networking, security threats, cloud computing and more. Network engineering teams will face increased pressure from more users, more networked devices and increased traffic to datacenters.
Manual processes will no longer be sufficient for network management, and, in particular, troubleshooting. The current network climate is demanding more scalable, flexible networks. Automated network diagrams will become the norm, and organizations that adopt this technology sooner will avoid the roadblocks and obstacles of manual troubleshooting that are currently costing organizations millions of dollars each year.
Grant Ho is a Senior Vice President at NetBrain Technologies. NetBrain is a leader in simplifying and reducing the efforts associated with network documentation, troubleshooting and change management by employing a tool everyone knows how to use – a map.