Effective management of a network's Layer 1 can positively impact the administration of Layers 2 through 7.
By Pat Thompson, TE Connectivity
Network managers in charge of Layer 1 understand the value of using physical layer management technology to document the network, but they need to convince information technology (IT) operations personnel of that value and how it integrates with their efforts. In this article, we'll look at several questions and answers for Layer 1 personnel to convey to their collaborators who oversee Layers 2-7.
Q:What is physical layer management?
A: Network managers want to have as much information as possible about the status of their networks. Traditional network management systems (NMS) or data center infrastructure management (DCIM) tools provide visibility into the configurations and status of routers, switches, and other equipment, but they don't reveal information about the state of the physical network. By reporting the status of fiber frames, patch panels, and which cable is connected to which port, physical layer management (PLM) systems provide the same visibility into Layer 1 of the network as an NMS or DCIM system does for Layers 2-7.
Physical layer management systems use intelligent fiber frames, patch panels, trunk cables, and patch cords to collect and report information about the physical state of the network. Unique chips at the ends of patch cords that are read by the ports on fiber frames and patch panels provide detailed information on what is being connected where, including port number, cable ID number, cable type, length, color, and media type. This information is reported to the NMS or DCIM that is used to document the physical layer.
When a cable is unplugged or plugged in, the system reports this in real time, so the network documentation is always up to date. This saves a lot of network technician labor that would otherwise be spent manually recording changes to the network.
Most physical layer management systems also include a work order management system that schedules tasks for network technicians and even guides them through a particular task. For example, an order to connect a circuit could trigger flashing green LEDs on the ports that are to be connected.
In addition, some work order management systems incorporate a web interface or mobile app for technicians. The mobile app can deliver requests for connection/disconnection tasks and then walk the technician through the process one step at a time. When the technician executes the tasks, the physical layer management system database is updated in near time to reflect this.
Q:How does physical layer management integrate with network management systems?
A: Network operations managers like to be able to see everything that's going on with a network through a single pane of glass. Network management software or DCIM software has its own physical layer database and its own work order management system. Using a separate software for the PLM system typically would lead to duplication of infrastructure asset databases and work orders. This can be avoided by integrating the existing management software with the PLM system using middleware integration.
With middleware integration, the PLM system's middleware application program interfaces (APIs) integrate with APIs from the NMS to convey physical layer information. The integrator will pick and choose which physical layer information is to be integrated by using simple GET commands.
The benefits of middleware integration are that the user doesn't have to implement and train people on a new management interface (for the PLM system alone), but can instead get the information needed through the management interface he or she is already using.
Middleware connects the PLM system to the NMS or DCIM and eliminates the need for a secondary database. Physical layer management middleware provides physical layer feedback and data in real time and is managed the same as Layer 2-7 information about the network.
Best practices in the software world have moved toward REST APIs and using web services languages like XML or HTML. Some physical layer management middleware systems support this model. Effectively, developers write scripts that connect APIs between the middleware and the target NMS or DCIM application. Suppliers supporting middleware are also working with standards bodies to standardize the data points that are available in a PLM system, so the APIs will be familiar and it's easier for all software to connect back to a PLM system.
In a script, basic commands like GET PORT STATUS can get an entire patch panel's status. When SNMP queries a device or the middleware, it will identify the configuration of the equipment through an SNMP Entity management information base (MIB). For example, the MIB will identify that Panel A Port 1 is connected to Panel B Port 12. It will automatically report all of its hardware components, and the configuration of those panels, and the configuration of each individual port. In all, there are only about half a dozen scripts that must be written to make the connection between middleware and an NMS or DCIM. This is generally not considered a major effort.
For integration with a work order management system, CONNECT commands tell the middleware which ports are involved, and to flash the green LEDs to tell the technician to connect a cable. The DISCONNECT command tells the middleware to flash the amber LEDs on the port to tell the technician to disconnect the port.
Q:How does physical layer management enable faster and more accurate service turnups?
A: There's a whole level of documentation that occurs prior to service activation. The documentation specifies which equipment goes where, where cables are run, and patch panel schedules, for example. In most cases, these details are maintained in spreadsheets or other manual entry systems, and the information is not always trustworthy.
Frequently, service activation begins with a technician going out to survey the network to verify that existing documentation is correct. Once verified, contractors and installers use these documents to get things installed. After the installation, technicians have to go back and test and verify that everything is connected properly. Without a PLM system, this is a manual, time-consuming process; one study shows that 70 percent of a technician's time is spent verifying the state of the network he or she is about to work on. Some organizations don't do surveys in advance. Instead, they just send the technician out with the work order, and the tech uses whatever port is available. The technician then has to manually reconcile the network documentation--an error-prone process.
With a PLM system, the network documentation is automatically updated in real time, so it can be trusted when planning new circuits, meaning operators can confirm very quickly whether or not it is possible to make the new connections. And as new circuits are installed, the PLM system automatically adjusts the documentation to reflect the changes. Service activations happen more quickly because there's no pre-installation survey needed, and technicians and network managers don't have to spend time manually updating documentation after the install or change is completed. Network administrators typically see a 50-percent improvement in the speed and activation of move/add/change (MAC) services.
Physical layer management systems also improve accuracy. When a technician goes to install a patch cord, he or she can be sure of the correct port because its green LED is flashing. If the circuit is to be disconnected, the amber LED flashes. These signals are sent out and managed by the PLM system middleware.
Q:How does physical layer management integrate with workforce processes?
A: A PLM system significantly streamlines workforce processes. In a typical network, technicians report to the work order or help desk to receive paper assignments, and then the technicians survey the network, make the change, and document the change on the paper work order. An administrator then takes the paper work order and updates the record-keeping system. With a PLM system, the process is automated; technicians can log into a website or use a mobile app to get assignments from wherever they are. They eliminate the network survey because they have confidence in the state of the documentation, and changes to the network are updated in real time as they happen.
Q: How does physical layer management save time and money across the network operations organization?
A: Without a PLM system, technicians spend most of their time verifying documentation or updating it. Technicians become significantly more productive when they have accurate documentation and don't have to worry about documenting changes. And over time, managed solutions offer more density and capital savings as unused ports can be reclaimed safely because of the full visibility and LED guidance.
Q:What's it like to install a physical layer management system? Is it disruptive?
A: Physical layer management systems consist of intelligent fiber frames, patch panels, and patch cords. These components look and function just like their "dumb" counterparts, so there's no need to retrain technicians. If the PLM system is being used only for automated documentation, the transition is completely seamless for network technicians. But if the goal is to implement automated work order management and task guidance, some minimal training will be required so technicians can learn to use the web interface and/or mobile app.
Q:Can a physical layer management system be installed on an existing network?
A: An existing network that has been operating for years, and whose documentation accuracy has degraded over time, will not be easy to convert to a managed system--which would require patch panels and cords to be replaced. Some solutions claim that they can retrofit onto existing installations. That really depends on the type of solution that is in place today and the type of physical layer management solution being deployed. A system with full intelligence and discoverability of connection details cannot be retrofitted.
Physical layer management systems are most effective in greenfield situations, or in network expansion projects, for example building all new extensions of an existing network with the management capability, such that over time the legacy installation can be decommissioned and the transition can be done gradually.
This is where middleware integration offers a benefit that the same management system and procedures can be applied to both the legacy portion of the network and the new managed sections of the network that will likely continue to coexist in a large network operation.
Q:What's the biggest objection to physical layer management?
A: Many organizations are not familiar with PLM solutions and their potential. When current processes are well understood by staff, they're not compelled to change when there's an increase in capital spending needed to make that shift.
When an organization does decide to migrate to physical layer management, it is done during data center expansions or new buildouts. Nobody expects an existing data center to rip out existing patch panels and cords and replace them with intelligent gear, but more and more data centers are putting technology in place to support this migration path today.
For those companies that do migrate to PLM, the return on investment is impressive. Hosted data centers typically see an ROI of six months, and independent data centers typically see an ROI of nine months.
By addressing these questions, Layer 1 network administrators can educate and reassure their Layer 2-7 counterparts that physical layer management is a big step forward.
Pat Thompson is director of global product management for TE Connectivity (www.te.com), responsible for physical layer management solutions. He has more than 20 years' experience in the telecommunications industry and has held a number of mechanical, design, and systems engineering positions, as well as product management and business development positions.