Category Archives: Network Layer

IP Classes

An Internet Protocol (IP) address is a logical address assigned to devices within a computer network. This numerical identification employs IP for transmission between nodes. IP addresses are in binary numbers but are usually shown in readable notations like 2001:db8:0:1234:0:567:1:1 (IPv6) and 192.168.100.1 (IPv4) .

IP addresses are primarily categorized into classes. The address class connotes the network’s potential size. The class of an IP address specifies the following:

  • bits that distinguish the network
  • bits that identify the network ID
  • bits that pinpoint the host ID
  • bits that isolate the host computer

It also demarcates the sum of host subnets in each network. There are five IP address classes: A, B, C, D, and E.

Host ID and Network Fields

Four octets that compose an IP address are classified under A, B, C, D, and E. The table below shows the distribution of the octets in classes A through C.

Class IP Address Network ID Host ID
A a.b.c.d a b.c.d
B a.b.c.d a.b c.d
C a.b.c.d a.b.c d

Class A – Addresses in this class are specified for networks with a large sum of hosts. Class A permits 126 networks and uses the first octet as the network ID.

Class B – Addresses in this class are specified for medium to large-sized networks. Class B permits 16,384 networks and uses the first and second octet as the network ID.

Class C – Addresses in this class are specified for small LANs (Local Area Networks). Class C permits approximately 2 million networks and uses the first, second, and third octet as the network ID.

QoS

Quality of Service (QoS), in traffic engineering, is a mechanism which prioritizes resource control over service quality in the field of computer networks as well as other packet-switched networks. Service quality is the capacity to assign different priorities to various applications, data flow, or users. It also guarantees data flow’s performance.

Service quality guarantees are useful if the network capacity for some applications may be insufficient, e.g. real-time streaming video, online games, etc. These services frequently need a fixed bit rate and are sensitive to delay, especially in networks of limited capacity. In networks where congestion is not a concern, QoS is not required.

A QoS-compatible protocol or network may have a traffic contract with its application software and resource capacity within the network nodes. During traffic management, QoS may observe the performance level (through data transmission rates and delay) and actively control schedule priority in the networks.

An alternate choice to complex QoS control systems would be the provision of high-quality communication over a network by ‘overestimating’ the network’s approximated capacity so that it will be more than sufficient for predicted peak traffic.

In telephony, Quality of Service is defined as a group of requisites based on the overall performance of one or more devices. QoS consists of requirements pertaining to all connection features, such as cross-talk, echo, frequency response, interruption, loudness levels, loss, signal-to-noise ratio, among other aspects. Under telephony QoS is Grade of Service (GoS), which consists of requirements for network capacity and coverage.

QoS may be utilized as a measure of quality rather than the ability for resource management. High QoS levels are often confused with high performance levels or the achieved quality of service (e.g. low bit error probability, low latency, and high bit rate).

Routing Protocols

A routing protocol is a set of standards that defines how routers communicate with each other. It enables routers to choose the path to utilize among a series of network within the Internet through a routing algorithm. In order to do that, it utilizes the following metrics in analyzing the path:

  • Count of the network layer devices along its path
  • Delay
  • Bandwidth
  • Cost
  • MTU
  • Load

Routing protocol is different from routed protocol. The former is responsible for determining the path on which to send traffic, the latter facilitates the actual sending of traffic or forwarding of packet from one host to another.

There are basically two types of routing protocols. They are:

  1. The Interior
  2. This type of routing protocols enables free dissemination of information within a given routing domain. The basic idea of the autonomous system (AS) lies in the fact that it allows the administrator independent control over the details within an AS. Hence, all the details of activity within one AS are made exclusive to that particular AS.

    Current interior routing protocols include the following:

    • Interior Gateway Routing Protocol
    • Enhanced Interior Gateway Routing Protocol
    • Routing Information Protocol (RIP)
    • The RIP automatically finds the routing tables on its own instead of the system administrator figuring them out. Hence, it utilizes the automatic routes and also redirects to a secondary path if the first one fails.

    • Open Shortest Path First
    • This is one of the most commonly used IGP’s. This, however, uses a different method in building route tables. It leans towards other OSPF enabled routers that have the same information used by other OSPF enabled routers in building its own route tables.

    • Intermediate System to Intermediate System
    • This type of routing protocol is similar to the OSPF. Under this are two types of addresses: the Network Service Access Point (NSAP) and the Network Entity Title (NET) .

  3. The Exterior
  4. This type of routing protocol facilitates routing information support autonomous system or outside a given domain. The following are common examples of exterior routing protocols:

    • Exterior Gateway Protocol
    • Constrained Shortest Path First
    • This particular algorithm is an advanced version of the SPF algorithm utilized for the OSPF and IS-IS route. When directing paths, the CSPF considers the following: topology of the network, attributes of the LSP, and the links.

    • Border Gateway Protocol (BGP)
    • The BGP is considered a distance vector protocol. Meanwhile, this type of protocol is also commonly used for the large internetworks and the Internet.

Traceroute

Traceroute pertains to a command that is used to mark out the route of a packet through a particular TCP/IP network. The Unix operating systems have the Traceroute command. Windows operating systems also use the Traceroute command, which has the name tracert.

Using the Traceroute command, individuals can determine certain pieces of information. Traceroute displays how many network layer hops had the packet passed through before reaching its destination. Hops refer to the transfers taken by the packet through the work stations on a network. This command also helps users to identify the other networks the packet would pass through to reach the station that requested it. Aside from this, Traceroute shows the number of times the packet crossed each network hop. Traceroute also shows the computer name of the host that responded to the packet request.

Generally, in order for an individual to use Traceroute, he should have any of these information: the destination computer’s IP address, name, or even website address.

To utilize the Traceroute command in Windows, the user first has to go to the command prompt. There he can enter “tracert (computer information)”. The quotes are not included, and the (computer information) should be replaced with either the computer’s IP address, website, or name.

Traceroute can also help in detecting subtle problems on the network. The user may see messages such as “request timed out” or the Traceroute command may display a message telling him that the web page he entered is loading slowly. In these events, the user can report the said messages to the network technician.

Static Route

A static route is one that is manually installed by your network administrator. This is a very efficient way to transfer data from one subnet to another despite the fact that this type of route is manually intensive.

Static route is a path in the router that indicates how it will reach a certain subnet by taking a specific path. This is called the hard coded path, which requires someone to input the network ID in order to reach the specified network.

Static Routes: The Basic

When sending traffic to a particular destination and you have two or more routes to choose from, administrative distance is utilized in determining which route to trust more. Hence, if you have entered a static and dynamic route, it will favor the static route. Nonetheless, you need to make sure that the routes are accurate while you do the manual overriding of routes.

A static route can either point towards the router’s interface or a network’s IP address. Therefore, you must determine which type of route to use variably.

Default route is a special kind of static route. This is often termed as “zero / zero” route because the destination to which it is sending traffic to, both subnet and network, are all in zero. Indeed, it catches any type of traffic that does not match any specific route in the routing table.

Another type of route is the dynamic route, which can be created through routing protocols. The dynamic route then chooses the best path to send traffic, unlike static routes, wherein the path to send traffic to are pre-determined.

Advantages of Static Routes

  • Static routes are easier to configure
  • No need for overhead on the routing protocol
  • As long as you have a tight IP mask, this offers you reliable security

Disadvantages of Static Routes

  • In order to make changes in the network, you have to manually configure the route
  • When network outage is experienced, it does not automatically route around
  • Although this is quite easy to configure, it might not work for large and complicated networks

It is important that any network administrator have substantial knowledge about static routes. Although this type of route may not be as effective with large networks, they are quite useful in any size of networks. Meanwhile, even if you have setup a dynamic route, there are cases that still require a static route.

Change an IP Address

An IP address is a numerical identification assigned to a computer to identify it within a network communication. Administrators usually designate these logical addresses based on network regulations.

IP addresses are changed for reasons of security and privacy, such as connecting to the Internet or a small network. Other Web users can also access your device’s IP address and can thus open access to your system. This access can be used for malicious reasons, such as denying admission to certain sites or services, or flooding your e-mail with spam.

Different operating systems use different steps in changing the IP address. In Windows 9X/ME, first go to Start, then to the Control Panel, then Network, click on the network card, next to the TCP/IP, and then change the IP address. When using Windows XP/2K, go to Start, then enter the Control Panel, then Network Connections. This will lead to Local Area Connection, the Properties, then to the Internet Protocol. Then change the TCP/IP. On the other hand, when Red Hat Linux I is being used, go to System Tools, followed by the Network OR use tools such as “linuxconf” or “netcfg” .

If these steps do not modify your IP address, call your Internet Service Provider (ISP) for suggestions. They will likely have special procedures regarding requests for IP changes. Your ISP may offer instructions on changing the IP address or give you new settings for future use.

Upon the approval of the request, modify your settings, and change the IP address to the new address provided by your ISP.

Note that if you modify the IP address to access Web-based forums you may choose to attempt to configure your Internet browser to use a proxy server.

For more information on Changing IP Address read:

  • Change IP Address
  • Change IP Address
  • Change IP Address
  • Change IP Address
  • 802.1Q

    802.1Q (or Virtual LAN) is a venture under IEEE (Institute of Electrical and Electronics Engineers) 802 standards. This project provides a mechanism with the capacity to bridge several networks to share a physical network link while prohibiting data leakage between them.

    802.1Q is also an interchangeable term used to refer to the standard issued by this project. It also pertains to procedures applied in the mechanism of Ethernet networks.

    As a standard, IEEE’s 802.1Q screens large networks into smaller components to minimize bandwidth use by broadcast and multicast traffic. This standard also contributes to a higher level of privacy between various internal segments within networks.

    The 802.1Q procedure provides a typical way of introducing VLAN membership data to Ethernet frames. In a Local Area Network (LAN), multicast and datalink-layer broadcast traffic transmitted to every end station. The traffic does not pass the LAN boundaries despite the fact that the perimeter shares cables or hubs.

    IEEE 802.1Q explains the meaning of VLAN based on the specific model bridging at MAC (Media Access Control) level and the 802.1D spanning tree protocol. This procedure allows individual VLAN communication using Layer-3 routers.

    If a company’s IT department wants to offer individual logical networks per specific department within the company while using one corporate network, VLAN will enable this route. Edge switches in the corporate network insert specific VLAN tags into the data frames coming from the equipment to and from any given department. The edge switches also removes the VLAN tags before the frames reach the specific department’s equipment (after switching off the frames in the corporate network). This ensures protection of one department from another.

    802.1p

    802.1p is an IEEE (Institute of Electronic and Electrical Engineers) standard that assigns traffic priorities while executing dynamic multicast filtering for other network mechanisms. 802.1p offers higher reliability and quality by prioritizing specification support, thus achieving higher Quality of Service (QoS).

    QoS is a system that endows better management of any data that passes through a network. There are two main classes of QoS – soft QoS and hard QoS. Soft QoS mainly involves data prioritization, while hard QoS deals with set capacities specific to certain types of service. QoS helps assure that integral data packets get to their destinations in the shortest span of time.

    The IEEE 802.1p standard assigns priority to packets crossing a network. The standard operates with the Media Access Control (MAC) header in the data link level. The MAC header is a portion monitored by a network’s switches and hubs. These devices also distinguish packets based on their priorities within the network.

    802.1p designates prioritization by setting a value within the MAC header. This value’s priority levels range from 0–7 (covering a sum of 8 levels), with level 0 denoting the lowest priority and level 7 being the highest. This allows packets to group and create various traffic classes. When network overcrowding happens, packets with low priority will be held temporarily while high-priority packets will be managed first.

    802.1p cannot work with older switches. Coexistence of the 802.1p with non-802.1p standards may lead to network instability. Older standards are bound to misread the header used by 802.1p protocol. For a network to operate properly, it is important that the device drivers, Ethernet cards, and the switches are compatible with 802.1p.

    ICMP

    The Internet Control Message Protocol (ICMP) is a core protocol of the Internet Protocol Suite. The Operating Systems in a computer network use this to send error messages, – for example, the unavailability of a requested service or the inaccessibility of a router or host.

    A network operation’s out-of-band messages use ICMP messages delivered through IP packets. Since ICMP utilizes IP, packet delivery is not wholly reliable, and hosts may not always obtain ICMP packets for all network concerns.

    ICMP announces network errors such as the inaccessibility of a host or the entire network portion due to any form of failure. ICMP reports if a UDP or TCP packet is targeted toward a port number without a receiver attached.

    It also announces network congestion. ICMP Source Quench messages are generated when a router buffers numerous packets. This is due inability of the transmission speed to keep up with the reception rate. These messages should slow down the speed at which packets are transmitted. Generating numerous Source Quench messages would cause network congestion, thus generation is moderated.

    ICMP also assists in troubleshooting. An Echo function sends a packet round-trip linking two hosts. A common network management tool named Ping centers on this. Ping measures average round-trips and computes loss percentages by transmitting packets.

    It also announces timeouts. The discarding router will often create an ICMP packet if the TTL field of an IP packet falls to zero. Traceroute, another network tool, maps routes through sending packets with minor TTL values and observes announcements of ICMP timeouts.

    Note that the Internet Control Messaging Protocol is a part of Internet Protocol; it is not foolproof. These control messages give feedback concerning communication problems in the environment. They do not make IP absolutely reliable. The ICMP messages usually report errors concerning datagrams, yet they do not guarantee the delivery of a datagram or return of a control message. In fact, datagrams may be undelivered and no report of this loss will be made. In addition, no ICMP messages regarding ICMP messages are sent to avoid the regress of messages on messages.

    If a reliable standard of communication is needed, then higher protocols using IP must apply their own particular reliability procedures.

    Routing Table

    A routing table is defined as an electronic document containing the various routes to all the nodes within a computer network. These nodes may refer to any type of electronic device connected to the physical network. The routing table is normally stored in a router or in a specific computer in the network. This document is in the form of a database file containing pieces of information organized in a particular structure.

    The main function of the routing table is to provide information and reference when a node in the network needs to transfer data to another node within the same network. Once the routing table is informed that a device will send data, the table searches for the best route for the data. After finding the best route, the table gives the device details regarding the route.

    Another more specific application of the routing table involves a process known as hop-by-hop routing. In this type of routing procedure, the address of each node leading to the destination is listed. The routing table is used by the data packet to determine the first node to go to from the source of the data. Once the data packet has reached that node, it will refer again to the routing table for the next node. This process will be repeated until the data packet has reached the target destination.

    Routing tables need to be consistent. Especially in large networks with many nodes and routers, the routing tables should have the same details and addresses contained in them. This ensures that data is safe from loops wherein they transfer from one node to another without reaching the intended destination.

    Collision Domain

    A collision domain refers to a logical segment in a network wherein the data packets being sent by devices collide with one another. The segment consists of devices that are connected via repeaters. One common protocol where collision domains can be seen is the Ethernet protocol.

    In an Ethernet environment, a collision domain is involved in the communication between nodes in the network. Hence, collision domains are often referred to as Ethernet segments.

    To avoid collisions, which lessen a network’s efficiency, the domain goes through a set of procedures in managing the nodes. When a particular device in the network sends a data packet to another device in the same network, the domain forces all other devices in the network segment to pay attention to the transfer. This is done until the data packet has reached its destination. Only one device may send data at a time.

    This set of procedures, however, does not fully prevent collisions. There may be certain instances where two devices attempt to simultaneously send data to the same destination. When this happens, the domain may not be able to prevent the two data packets from colliding. The target node will then stop the transmission and give a jam signal informing the other nodes that a collision took place. The two devices will subsequently yield and send their data at another time.

    Collision domains are associated with a number of devices, such as switches and routers. These pieces of hardware enable two or more collision domains to communicate with one another.

    Broadcast Domain

    A broadcast domain refers to a logical part of a network, in which any equipment within the network can directly send data to another equipment or device. When in the broadcast domain, data transfer can be done without going through a routing device. Before this can be done, however, the devices are required to share one subnet and the same gateway. They should also be in the same Virtual Local Area Network or VLAN.

    Basically, a broadcast domain is reached by sending a frame to the broadcast address of the data link layer. The frame being sent to a specific destination can be detected by all the devices in that location, but only the device to which the frame is addressed will be able to receive it. The interconnections in broadcast domains can only be divided by devices under the layer 3 network, examples of which are layer 3 switches and routers.

    There are a few cases wherein broadcast domains need to be restricted. This is done to increase the level of security, and to prevent the devices from receiving unwanted data and from carrying out unauthorized tasks. The restriction of broadcast domains is done by routers. A router has the option to drop some of the broadcast signals it receives. Certain issues are raised regarding this restriction, however, especially in cases where a network requests information from another network via broadcast signals.

    The use of the broadcast domain is employed in a number of circumstances. Workstations announcing requests by indicating their location in the network apply the procedures invoking the broadcast domain. Network devices sending data from a single LAN segment to another also make use of the broadcast domain.