Putting together a 24-hour, 7 days a week support system is expensive.
The software alone will not magically improve performance. As your company grows, hiring an in-house network technician will become a necessity.
To hire multiple technicians with the appropriate skills and training at roughly $30 per hour, 24-hours per day will cost about $720.00 each day. That’s $5,040 per week or $20,160 per month for 24/7 coverage.
In this hypothetical situation, those salaries will cost your company $262,800 on the year, plus the cost of the software and infrastructure.
When the costs are realized, it’s the primary reason why companies opt to outsource with a trusted network operations center (NOC) rather than tackle network monitoring internally.
If your company has come to this conclusion, you may have decided on NOC services to support your team and you’re wondering about the onboarding process.
CCI Systems has a team of experts who monitor big and small networks 24 hours a day, 7 days a week, 365 days a year. Our support services team has onboarded NOC customers in less than 10 days!
Plus, CCI is uniquely positioned to use your existing monitoring system without skipping a beat.
The NOC onboarding process starts with identifying how you want to monitor your network, what devices need to be monitored, creating a knowledge base of policies and procedures, going for a test drive with a proof of concept, and after the go-live date, deciding what continued support might look like for your company.
To kick off a discovery call, a company needs to decide how they want to monitor their network. There are three basic solutions to consider:
The software solution is the most straightforward option.
All the required devices from routers to switches are hooked into the networking monitoring software. This can be existing software the company uses to monitor their network, or it can be new software recommended by the NOC services provider.
Some options of network monitoring software are Solarwinds, LogicMonitor, Datadog, Nagios, and Dynatrace.
Once the data stream is functional, the network and the network’s health will be visible through interactive dashboards, charts, and reporting tools.
It will be the responsibility of the company and their staff to master the software, not the NOC.
This solution comes after the prerequisite of monitoring software is in place, and it has been decided the NOC will monitor all incoming alerts for the network.
Per the process and procedures designated by the client, the NOC will receive the alert and forward it to the company or a specific contact at the company, or the NOC will be given permission(s) to troubleshoot.
Identifying what alerts matter and creating levels of prioritization will help to ensure the correct actions are being taken for each alert.
For instance, there will be times where a note for a technician in the morning will suffice. The NOC will identify an alert as a low priority, and the company will be sent a note to check on the ticket at their earliest convenience.
If the alert is a high priority (i.e. a network outage), the NOC will begin to troubleshoot and alert the client immediately.
Good alert management will allow for false notifications generated by the software solution to be weeded out before a truck roll takes place.
The most common type of network monitoring is a turnkey solution.
A turnkey solution includes the software installation and setup, and the alert management aspect, but goes a step further to include the option of corrective action in the event of an outage.
Typically, your company will want the support services team you’ve hired to do most of the heavy lifting and work, like working with your engineers for break/fix and incident response.
Once the network monitoring solution is chosen by the client, the type of equipment and the number of devices will be cataloged.
When you’re talking equipment, it’s always smart to think big-to-little. If you take a rural cable company, for example, you’ll find a common network design.
First, a core router will be put at each head end. Underneath those core routers are transports.
Within a transport, five fibers go in, the information is fed across a single fiber that's going out, and then, it splits the information back up through the output. This process is also called a demultiplexer (or demux).
Underneath the transport are the core switches. The transport gets the data, and the router determines where it's supposed to go. This data is then sent down to a switch connected to five customers (or end users).
At each of those five locations, there may be another switch or smaller router before being dispersed to the customer network. It’s very straightforward.
When considering the design of your network, know that some NOC providers will charge by the port for monitoring services, while others will charge by the piece of equipment.
Being aware of the “per port” versus “per device” billing will help you get the best pricing.
After this is equipment check is completed, one of two things will happen.
In the proof of concept, the NOC will hook into select devices, set up the appropriate dashboards, show them what it's like to see their network data in a single window. It’s like a taste test at a 5-star restaurant.
Typically, a trial period lasts from 30 to 60 days.
Some companies may need an immediate solution, and they skip the trial run and test drive. They might say, “Let's take our core devices, and we’ll at least have you monitor them. Let's start with that.”
In this scenario, the NOC may jump right into a project plan to onboard those devices, which can occur in less than 30 days.
Either decision requires network documentation, an equipment list, and finally, hooks into the data source. The data source is typically simple network management protocol (SNMP) data.
With a cloud-based monitoring system, like Solarwinds or Nagios, the solution provider needs access to a virtual server to connect, comparable to logging into a secure VPN that is behind a firewall.
That receives SNMP traffic from their device and sends information over to a dashboard. When there is a network discrepancy, an alert is sent to whoever is monitoring.
As a provider, knowing a third party will be in your network with cloud-based monitoring capabilities can make the hair on the back of your neck stand up.
Because of this potential vulnerability, it is extremely important to find a network operations center practicing safe and secure methods for monitoring. This is to ensure the data stream or API where the information is being delivered is unalterable.
If this is not spelled out and covered in perpetuity, it is possible an irresponsible NOC without a focus on security could cause a data breach.
For a solution provider, this presents a huge, huge obstacle. Some companies think if they buy the software, and they monitor it, that they're more secure. And that is the fallacy.
Network surveillance from the NOC down to the company providing the network monitoring software must be secure.
After the alert comes through, the monitoring software has done what it’s supposed to do. That's the monitoring process in a nutshell.
Once the alert is received, the support process kicks in. The NOC’s job is to identify what to do with this information with the knowledge that a piece of equipment has taken a hit.
A few examples of alerts are a piece of equipment that is not running at peak performance, part of a shelf died, a port serving a section of a company’s customer base is down, or perhaps, the whole network went down.
This is where a thorough list of processes and procedures comes come in. It's either notify and alert (i.e, “Hey Dave, your device is down. Have a nice day.”) or notify and begin troubleshooting on the customer’s behalf.
When the NOC is operating with a turnkey solution, a technician will see a device is down, send a ticket to let the provider, and find out what the actual issue is. It may require the NOC to begin an escalation process, and they will ask for a technician to be sent to the alert’s location.
The onsite technician or engineer will then work directly with the NOC until the device is powered up, resumes communication with the NOC, and the communication is verified.
In some special cases, if it's a circuit between Site A and Site B, the circuit may be provided by a Tier 1 provider who in turn providers the technician(s) in the case of a truck roll.
In this situation, the NOC opens a ticket on behalf of the client with their Tier 1 provider. Then, the NOC works with the onsite technicians or engineers.
Strategically choosing what devices your company will monitor can make a big difference in seeing the full breadth of your network’s health. In the end, a clear view of the network can lead to better customer service.
Based on the equipment and applications, a knowledge base is built. This knowledge base covers all NOC permissions, it is customized to each company being onboarded, and it is the textbook to train the NOC employees.
Each piece of equipment and application will come with its own set of instructions for the NOC to follow. The map is developed by the NOC and the company, or the company will come with its own set of policies and procedures.
*All policies and procedures are customizable to each customer.
Some companies will require the alert to be sent to them first. Some will only want the NOC to take two steps to correct an issue. Others will be hands-off and ask for the entire process to be handled by the NOC.
For example, a NOC may be monitoring a network and there is a fiber node serving five hundred end users. If that node sends the NOC an alert message that it's in distress, a company may explicitly state they want their technicians notified immediately because it is to be handled internally.
Conversely, the company may not have the resources or technicians to address the alert, so they will allow the NOC technician(s) to access the application and begin troubleshooting.
Troubleshooting can involve rebooting the node, seeing if the fiber has been cut, or simply running diagnostics.
NOC technicians become part of the team.
Referred to as a proof of concept, a “test drive” rolls off the tongue a little easier.
At this point in the onboarding process, each device has been accounted for and prioritized, a knowledge base has been completed, the NOC technicians have been trained on those policies and procedures. Finally, the NOC begins monitoring.
It’s very similar to Day 1 of the process in that a call will happen at the end of each day for the first week. Questions will be asked about tickets, critiques will be made on the troubleshooting front, and changes will be made to accommodate any situations left unaccounted for while building the knowledge base.
A week of daily calls will be proceeded by a call once per week for the next four weeks. After that, if the network monitoring process is going as planned, there will be a shift to one call per month.
In some cases, companies will find quarterly calls are sufficient. The ability to see each ticket placed for any alert coming through the network operations center is oftentimes enough transparency for a service provider. Plus, they will have access to dashboards providing them with every insight they need.
No customer is exactly the same. No network is exactly the same.
That's why making sure all documentation is in order and having all escalation contacts set up in advance are critical pieces to cover within the onboarding process.
Ultimately, a sound and complete knowledge base will determine when your company is ready to go live.
Once your company has gone live, both parties will see the data stream together.
Both the provider and NOC will have mutually agreed that the dashboards are accurate and correct, the documentation is complete, and the network data is flowing securely through the system.
With each of those boxes checked, the onboarding process will be complete.
After weeks of back and forth where both parties are verifying the correct data is being seen, the configuration and the timing of alerts are as good as they can get, and the speed of communication between the NOC and provider is determined, there will still be ongoing variables to address as the relationship unfolds.
A common variable for providers is the network receiving a fair share of frequent, false alerts. Sometimes, this may be a result of all maintenance being completed between midnight and 6:00 A.M. but the NOC wasn’t notified. To accommodate this change, the NOC must adjust the procedures within alert management to ignore those notifications triggered during those maintenance windows.
This is an ode to the extensive coordination and collaborative effort between the NOC services team and the company seeking the service.
Equipment can always be added to NOC services. Adding more equipment to network monitoring is a common practice as network upgrades or rebuilds occur.
After a year of monitoring core devices, a company may want to begin monitoring non-core devices with potential vulnerabilities. The contract may cover 60 devices to be monitored, but the following quarter, the number increases to 80 devices after a network expansion. This is a very common occurrence.
Whether more devices are eventually added down the road, largely depends on the size of a company and the saturation of the total addressable market.
The purpose of a network operations center is to add another layer of protection to maintain the health of your network. Plus, it is more cost-effective to outsource NOC services than it is to take on the network monitoring in-house.
Onboarding will define how each protection is developed, implemented, and executed during an incident. Then, proactive network monitoring will help mitigate the occurrence of those incidents.
CCI Systems provides 24/7 network monitoring services to providers who are just starting and others who have been around for decades. CCI’s NOC is SOC 2 Type II-certified with a yearly audit to ensure client data is as safe and secure as it can be.
Concerned about outsourcing network monitoring? Here are some answers to questions and concerns providers frequently bring to the table, and the reasons why your company might want to reconsider.