CompTIA Network+ N10-008 – Module: Examining Best Practices for Network Administration Part 2

3. 16.2 Wiring Management

In this video, let’s talk about some best practices surrounding wiring management. First, let’s define a couple of terms. We have intermediate distribution frames or IDFs, and they might be scattered around a large building. We might call these a wiring closet. It might be some equipment room where there’s a large piece of plywood mounted on the wall and it’s into these IDFs, these intermediate distribution frames to which our end devices connect. We’ve got PCs or routers or switches, any sort of wired networking equipment that’s going to be going back oftentimes to the closest to IDF.

And these IDFs, instead of having a full mesh of connectivity between them, what we typically do is send all of the IDFs back to an MDF, a main distribution frame like you see on screen. This way we can get from any single IDF to any other IDF. And when you’re looking at an IDF, you might see a piece of plywood on the wall, and mounted to that plywood you might see a 110 block or maybe a patch panel. In either case, we have a punch down tool to punch down the wires coming from the PCs and the devices out in the office area.

And that’s if we’re dealing with copper cabling. But if you have fiber coming into one of these IDFs or an MDF, a caution about working with fiber cable if you recall how fiber works, we have a couple of different types of Glyce. In a fiber optic strand, there is a core and there is a cladding. And the core and the cladding have different indices of refraction. Light travels at different speeds between the glass and the core and the glass in the cladding. And when a light wave inside of the core hits the cladding, it’s supposed to bounce back into the core based on that different index of refraction.

But that optical principle does not work. If we bend the fiber too tightly, if we bend it too tightly or we wind it up with a radius that’s too small, then that light is going to escape the core and go into the cladding and degrade our signal. For example, let’s say that I have a seven foot jumper cable and I plug one end into my equipment and I need to plug the other end into a fiber optic patch panel. And after I do that, I have a few feet of fiber just hanging there.

Well, I don’t want it to be a mess. So what some people will do is fold it up and put a couple of cable ties around it, making it very tight. It looks good. But think about what they’re doing. If they’re bending that fiber too tightly, then suddenly they’re losing light. When they folded that cable up and secured it with a cable tie, the bend radius of that fiber is way too small and you can have significant signal loss. If that fiber is bent too tightly, the light is escaping from the core and going into the cladding. And oftentimes a fiber patch panel will have holders where you can safely loop extra fiber cabling around in a circle without going below that minimum bend radius where you start to lose light. Let’s think about a couple of other wiring management best practices. One deals with the kind of copper cabling that we might use in a raised floor or any sort of an open air duct.

Any place you have air flowing, you want to use plenum cabling. Think about this. Let’s say that we’ve got some catsix wiring going through the ceiling. And the way that the HVAC system is set up is we’re blowing air through that ceiling area or maybe through a raised floor in a data center. What if there’s a fire? Well, if there’s a fire, if we’re using non plenum cabling, that insulation around the copper cabling, if it’s exposed to fire or a lot of heat, it could start to melt and release toxic fumes. However, plenum cabling, which is the kind of cabling we should use in those situations, in those plenum areas, as they’re called, its outer coating, is not only fire retardant, but it’s not going to release those toxic gasses if exposed to fire or heat. So please use plenty of cabling, any place you might have airflow from an HVAC system, because you don’t want to be blowing toxic fumes into other areas of the building.

And one of the biggest things we try to avoid when it comes to wiring management is we want to avoid what is often referred to as a spaghetti wiring. Have you seen this? You go into a wiring closet or you look at a patch panel, and the cables are just all over the place. They’re tangled together, they’re piled up on the floor. If you’re trying to find what cable goes to port gig zero on this switch, it’s difficult to find in that mess of spaghetti wiring. What are some ways that we can make things a bit more organized and it doesn’t look like a big plate of spaghetti on the floor? Well, for one thing, let’s make sure that we’re using correct cable lens. Oftentimes, if we need to interconnect a couple of pieces of equipment and they’re in the same rack, we may only need something like a 1ft jumper. We might say, well, I don’t have a 1ft jumper, but I’ve got this 25 foot cable that’s going to leave a lot of excess cable just lying there. So let’s take the extra step of making sure that we have correct cable lengths and in racks and in desks. Let’s use any cable management systems that are in place.

A lot of times between racks and a data center, there will be cable trays. Interconnecting the racks. Let’s make sure we put the cable in those cable trays. And as a best practice, we do not want to have our data cabling lying in the same cable tray as Power Cabling. We don’t want the electricity going through the power cable to cause any sort of interference with our data cable. Let’s separate those. And let’s also do a good job of labeling both ends of a cable. You might be looking at a cable that’s plugged into a switch that’s mounted in a rack, and you’re trying to figure out where this goes over on this 110 block. Well, I suppose you could use one of those fox and hand devices and you could figure out where it’s terminating on the 110 block. But to do that, you would have to disconnect the cable from the switch. It would be great if you could figure out where this wire is appearing on the 110 block.

And we can do that simply by labeling it as we’re doing the installation. Let’s label both ends of the cable, and that’s a collection of wiring management to best practices that we need to think about. So to sum up in this video, we defined a couple of terms. An intermediate distribution frame or an IDF. We could have several of those in the building, and those IDFs, they connect back to the building’s MDF, the main distribution frame. And in those IDFs, we talked about having 110 blocks or patch panels there, and that’s for Copper Cabling. If we have fiber cabling. We talked about not going below the minimum bend radius. And we talked about the importance of using plenum cable where we’re not releasing toxic fumes into open air return ducts. And we also talked about some best practices for avoiding spaghetti wiring, having correct cable lens, using any cable management systems that are in place, and labeling both ends of a cable.

4. 16.3 Power Management

Our network gear cannot function without power. And in this video we want to take a look at some power options. For example, let’s say that we have a server, maybe in a data center. It could just be plugged into the wall. It could be receiving AC power from the building’s power. However, what if there is a power outage? I mean, that can occur from time to time. Maybe weather conditions knock out the power. What we can do is have that server plugged not into a wall outlet, but into a Ups, an uninterruptible power supply. And with a Ups, the power is coming from a wall outlet into the Ups and then into the server. There’s a battery inside of the Ups that’s sitting in line with the server, meaning that if power is lost, there is no cut over time, the server is immediately receiving power from that battery inside of the Ups. And that’s a distinction between a Ups and an SPS. An SPS might be a lower cost option as you’re doing your shopping, it might be a lower cost option to a Ups. An SPS, that’s a standby power supply.

And with a standby power supply, there is a battery, but it’s not in line with the server. When a power outage is detected, there’s a mechanical relay that puts the battery in line with the server. And it might take several milliseconds to do that. That might not be acceptable to us or tolerated by our equipment. If you truly want to have no uninterrupted power, you might want to consider a Ups. And we said that a Ups has a battery inside of it, but a battery that has a finite life, what do you do when the battery runs down? Well, what many people do is have a generator in addition to a Ups. I remember when I used to work at a university, we had a fairly large server farm and we had several Ups within that server farm.

And sure enough, we would occasionally have a power outage. What would happen? Well, when the power outage occurred, the Ups is they would power the servers using their battery for a time and that would give the generator time to get going. It would generally take about 45 seconds or so after the power outage occurred for the generator to kick on. But once the generator kicked on, it could start providing the power.

As long as we had gasoline in the generator, it could provide power to the servers, and we were able to go for hours and hours without building power, keeping that server farm up and going. So when it comes to power redundancy, think about having a Ups, but also think about what’s going to happen if you have an extended power outage. Do you have a generator to take over for the Ups? And speaking of power redundancy, let’s think about a data center for a moment. A lot of our high end equipment. And here we’re looking at a Cisco catalyst to switch.

But a lot of our high end equipment has a couple of different power supplies. Now that’s great for redundancy. If one power supply goes out, we have another power supply that can take over. And what we can have in some larger data centers is we can have a couple of PDUs. Those are power distribution units. And if it’s available, some data centers have these power distribution units serviced from different providers. One provider might have an outage, but the other provider does not.

And what you oftentimes see in a data center is that the two power supplies in a piece of equipment, they’ll be connected to different PDUs. So if one power distribution unit goes down, the other power distribution unit is hopefully still up and it can continue to provide power to the equipment. And one other thing I want you to think about when you’re thinking about providing power to your equipment, I want you to think about the kind of power you need to provide. Remember that voltage might vary country to country. The outlets might vary country to country. And the amount of current that has to be carried by the electrical circuit, that might dictate what our outlet looks like. And also make sure you’re providing the appropriate type of electricity to your equipment.

There is AC, or alternating current, and there’s DC direct current. With alternating current, if you look at it on an oscilloscope, it looks a lot like a sine wave where the voltage goes up and down and up and down. It alternates. But with direct current, it just has a steady flat voltage level. And make sure you know what your network device needs.

And I took a picture of the back of one of my switches and you can see that it has an option for DC input. Normally you’ll just plug a power cable, you would plug it into a wall or into a Ups, and that’s going to power the switch. But here there’s an option for direct current. And I might want to use this option if I’m on maybe a ship or a military vehicle that has direct current but does not have alternating current available. And those are some power management considerations.

• Category: Uncategorized