When you're getting started on your CCNA studies on your way to earning this certification, you're swamped with network device types that you're familiar with, but not quite sure how to use. Let's look at these networking devices and their main purposes.
Hubs and repeaters operate at Layer One of the OSI model, and they have one main purpose - regenerating the electrical signal that Layer One technologies carry. This regeneration helps to avoid attenuation, the gradual weakening of a signal. Much like a radio signal, the electric signals that travel at Layer One gradually weaken as they travel across the wire. Hubs and repeaters both generate a "clean" copy of the signal.
While hubs and repeaters can be helpful, they do nothing as far as network segmentation is concerned. The first such device we encounter as we move up the OSI model is the switch. Operating at Layer 2, a switch creates multiple collision domains by default each switch port is considered its own little collision domain. If 12 PCs are connected to a Cisco switch, you have 12 separate collision domains.
Switches can be used to segment the network into smaller broadcast domains, but this is not a default behavior. Virtual LAN (VLAN) configuration segments the network into smaller broadcast domains, since a broadcast sent by a host in one VLAN is heard only by other devices in the same VLAN.
Routers operate at Layer 3 of the OSI model and segment a network into multiple broadcast domains by default. Routers do not forward broadcasts as switches do, making the router the only device of the four we've discussed today that create multiple broadcast domains by default.
Knowing what each of these devices can and cannot do is essential to passing the CCNA and becoming a great network administrator. Good luck to you in both of these goals!
Showing posts with label collision. Show all posts
Showing posts with label collision. Show all posts
Wednesday, December 24, 2008
Cisco CCNA Certification: Defining Broadcast Domains
When you're studying to pass the CCNA exam and earn your certification, you're introduced to a great many terms that are either totally new to you or seem familiar, but you're not quite sure what they are. The term "broadcast domain" falls into the latter category for many CCNA candidates.
A broadcast domain is simply the group of end hosts that will receive a broadcast sent out by a given host. For example, if there are ten host devices connected to a switch and one of them sends a broadcast, the other nine devices will receive the broadcast. All of those devices are in the same broadcast domain.
Of course, we probably don't want every device in a network receiving every single broadcast sent out by any other device in the network! This is why we need to know what devices can create multiple, smaller broadcast domains. Doing so allows us to limit the broadcasts traveling around our network - and you might be surprised how much traffic on some networks consists of unnecessary broadcasts.
Using the OSI model, we find devices such as hubs and repeaters at Layer One. This is the Physical layer, and devices at this layer have no effect on broadcast domains.
At Layer Two, we've got switches and bridges. By default, a switch has no effect on broadcast domains; CCNA candidates know that a switch will forward a broadcast out every single port on that switch except the one upon which it was received. However, Cisco switches allow the creation of Virtual Local Area Networks, or VLANs, that are logical segments of the network. A broadcast sent by one host in a VLAN will not be forwarded out every other port on the switch. That broadcast will be forwarded only out ports that are members of the same VLAN as the host device that sent it.
The good news is that broadcast traffic will not be forwarded between VLANs. The bad news is that no inter-VLAN traffic at all is allowed by default! You may actually want this in some cases, but generally you're going to want inter-VLAN traffic. This requires the use of a router or other Layer 3 device such as a Layer 3 Switch. (Layer 3 Switches are becoming more popular every day. Basically, it's a switch that can also run routing protocols. These switches are not tested on the CCNA exam.)
That router we just talked about also defines broadcast domains. Routers do not forward broadcasts, so broadcast domains are defined by routers with no additional configuration.
Knowing how broadcasts travel across your network, and how they can be controlled, is an important part of being a CCNA and of being a superior network administrator. Best of luck to you in both of these pursuits!
A broadcast domain is simply the group of end hosts that will receive a broadcast sent out by a given host. For example, if there are ten host devices connected to a switch and one of them sends a broadcast, the other nine devices will receive the broadcast. All of those devices are in the same broadcast domain.
Of course, we probably don't want every device in a network receiving every single broadcast sent out by any other device in the network! This is why we need to know what devices can create multiple, smaller broadcast domains. Doing so allows us to limit the broadcasts traveling around our network - and you might be surprised how much traffic on some networks consists of unnecessary broadcasts.
Using the OSI model, we find devices such as hubs and repeaters at Layer One. This is the Physical layer, and devices at this layer have no effect on broadcast domains.
At Layer Two, we've got switches and bridges. By default, a switch has no effect on broadcast domains; CCNA candidates know that a switch will forward a broadcast out every single port on that switch except the one upon which it was received. However, Cisco switches allow the creation of Virtual Local Area Networks, or VLANs, that are logical segments of the network. A broadcast sent by one host in a VLAN will not be forwarded out every other port on the switch. That broadcast will be forwarded only out ports that are members of the same VLAN as the host device that sent it.
The good news is that broadcast traffic will not be forwarded between VLANs. The bad news is that no inter-VLAN traffic at all is allowed by default! You may actually want this in some cases, but generally you're going to want inter-VLAN traffic. This requires the use of a router or other Layer 3 device such as a Layer 3 Switch. (Layer 3 Switches are becoming more popular every day. Basically, it's a switch that can also run routing protocols. These switches are not tested on the CCNA exam.)
That router we just talked about also defines broadcast domains. Routers do not forward broadcasts, so broadcast domains are defined by routers with no additional configuration.
Knowing how broadcasts travel across your network, and how they can be controlled, is an important part of being a CCNA and of being a superior network administrator. Best of luck to you in both of these pursuits!
Cisco CCNA Certification: Defining And Creating Collision Domains
When you're studying to pass the CCNA exam and earn your certification, you're introduced to a great many terms that are either totally new to you or seem familiar, but you're not quite sure what they are. The term "collision domain" falls into the latter category for many CCNA candidates.
What exactly is "colliding" in the first place, and why do we care? It's the data that is being sent out onto an Ethernet segment that we're concerned with here. Ethernet uses Carrier Sense Multiple Access / Collision Detection (CSMA/CD) to avoid collisions in the first place. CSMA/CD is a set of rules dictating when hosts on an Ethernet segment can and cannot transmit data. Basically, a host that wants to transmit data will "listen" to the ethernet segment to see if another host is currently transmitting. If no one else is transmitting, the host will go forward with its own transmission.
This is an effective way of avoiding a collision, but it is not foolproof. If two hosts follow this procedure at the exact same time, their transmissions will collide on the Ethernet segment and both transmissions will become unusable. The hosts that sent those two transmissions will then send a jam signal out onto the segment, indicating to all other hosts that they should not send data. The two hosts will each start a random timer, and at the end of that time each host will begin the listening process again.
Now that we know what a collision is, and what CSMA/CD is, we need to be able to define a collision domain. A collision domain is any area where a collision can theoretically take place, so only one device can transmit at a time in a collision domain.
In another free CCNA certification tutorial, we saw that broadcast domains were defined by routers (default) and switches if VLANs have been defined. Hubs and repeaters did nothing to define broadcast domains. Well, they don't do anything here, either. Hubs and repeaters do not define collision domains.
Switches do, however. A Cisco switchport is actually its own unshared collision domain! Therefore, if we have 20 host devices connected to separate switchports, we have 20 collision domains. All 20 devices can transmit simultaneously with no danger of collisions. Compare this to hubs and repeaters - if you have five devices connected to a single hub, you still have one large collision domain, and only one device at a time can transmit.
Mastering the definition and creation of collision domains and broadcast domains is an important step toward earning your CCNA and becoming an effective network administrator. Best of luck to you in both these worthwhile pursuits!
What exactly is "colliding" in the first place, and why do we care? It's the data that is being sent out onto an Ethernet segment that we're concerned with here. Ethernet uses Carrier Sense Multiple Access / Collision Detection (CSMA/CD) to avoid collisions in the first place. CSMA/CD is a set of rules dictating when hosts on an Ethernet segment can and cannot transmit data. Basically, a host that wants to transmit data will "listen" to the ethernet segment to see if another host is currently transmitting. If no one else is transmitting, the host will go forward with its own transmission.
This is an effective way of avoiding a collision, but it is not foolproof. If two hosts follow this procedure at the exact same time, their transmissions will collide on the Ethernet segment and both transmissions will become unusable. The hosts that sent those two transmissions will then send a jam signal out onto the segment, indicating to all other hosts that they should not send data. The two hosts will each start a random timer, and at the end of that time each host will begin the listening process again.
Now that we know what a collision is, and what CSMA/CD is, we need to be able to define a collision domain. A collision domain is any area where a collision can theoretically take place, so only one device can transmit at a time in a collision domain.
In another free CCNA certification tutorial, we saw that broadcast domains were defined by routers (default) and switches if VLANs have been defined. Hubs and repeaters did nothing to define broadcast domains. Well, they don't do anything here, either. Hubs and repeaters do not define collision domains.
Switches do, however. A Cisco switchport is actually its own unshared collision domain! Therefore, if we have 20 host devices connected to separate switchports, we have 20 collision domains. All 20 devices can transmit simultaneously with no danger of collisions. Compare this to hubs and repeaters - if you have five devices connected to a single hub, you still have one large collision domain, and only one device at a time can transmit.
Mastering the definition and creation of collision domains and broadcast domains is an important step toward earning your CCNA and becoming an effective network administrator. Best of luck to you in both these worthwhile pursuits!
Cisco CCNA Certification: Cisco Switching Modes Tutorial
To pass the CCNA exam and earn that coveted certification, you've got to know Cisco switches inside and out. Among the many important details you've got to know are the three methods that Cisco switches use to forward frames, and the differences between the three.
The first switching method is Store-and-Forward. The name is the recipe, because that's just what the switch does - it stores the entire frame before beginning to forward it. This method allows for the greatest amount of error checking, since the Frame Check Sequence (FCS) can be run before the frame is forwarded. As always, there is a tradeoff, since this error checking process makes this the slowest of the three frame forwarding methods.
The quickest method is Cut-Through, where only the destination MAC address of the frame is examined before the forwarding process begins. This means that the part of the frame is actually being forwarded as it is still being received! The tradeoff here is that the FCS does not run, so there is absolutely no error checking with Cut-Through switching.
The middle ground between these two extremes is Fragment-Free, so named since fragmented frames will not be forwarded. The switch examines only the first 64 bytes of the frame for errors, since that is the part of the frame that will be damaged in case of a collision. There is error checking, but it is not as thorough as Store-and-Forward.
Keeping these three switching schemes straight is vital to your CCNA exam efforts, and it will help you in working with Cisco switches in the real world as well. Keep studying!
The first switching method is Store-and-Forward. The name is the recipe, because that's just what the switch does - it stores the entire frame before beginning to forward it. This method allows for the greatest amount of error checking, since the Frame Check Sequence (FCS) can be run before the frame is forwarded. As always, there is a tradeoff, since this error checking process makes this the slowest of the three frame forwarding methods.
The quickest method is Cut-Through, where only the destination MAC address of the frame is examined before the forwarding process begins. This means that the part of the frame is actually being forwarded as it is still being received! The tradeoff here is that the FCS does not run, so there is absolutely no error checking with Cut-Through switching.
The middle ground between these two extremes is Fragment-Free, so named since fragmented frames will not be forwarded. The switch examines only the first 64 bytes of the frame for errors, since that is the part of the frame that will be damaged in case of a collision. There is error checking, but it is not as thorough as Store-and-Forward.
Keeping these three switching schemes straight is vital to your CCNA exam efforts, and it will help you in working with Cisco switches in the real world as well. Keep studying!
Subscribe to:
Posts (Atom)
