|
Network OverviewThe following section is an overview of computer network technology as used in the DCX system. The purpose of this section is to familiarize you with the background information necessary to configure the networks setting of the DCX and client workstations. It is not intended to be used as a thorough reference on the subject of computer networks. For more detailed information consult a network professional or one of the many books available on the subject. Your DCX audio system uses TCP/IP networking to provide communications between the client workstations and the DCX control rack. Each network device or node in a TCP/IP system must be assigned at least one unique 32-bit Internet Protocol (IP) address. These addresses make it possible for the client computers and DCX to send control messages to each other, much in the same way that individual telephones have their own phone number. Depending on whether you will be connecting you DCX system to an existing network or creating a new one, you will need to choose an appropriate set of node addresses and assign one to each DCX system component. For connection to an existing network system, you will need to contact your network administrator to assist in choosing the proper addressing scheme. The DCX system uses the built-in network features of Microsoft Windows to provide TCP/IP network support. Before the network features of the DCX system can be used for the first time, a network interface card (NIC) must be installed in each client computer and the proper network components must be loaded and configured for each computer. Your DCX control rack comes with a network interface card already installed and the network software ready to run for new installations. If you are connecting to an existing network system, or if you need to use a custom set-up, you will need to change the network settings on the DCX control rack. A basic DCX control system is shown below. Here a DCX control rack in connected to two client workstations on an Ethernet network. The DCX and each workstation are outfitted with Ethernet network interface cards. The cards from each device are wired to a network hub that provides proper cable termination and signal buffering between the connected devices. The cabling used in this system conforms to the IEEE 10Base-T standard for interconnecting network devices. The hub device and cabling are common items and are readily available at most computer and electronics stores. Variations of this system can include other network components and different cabling schemes such as fiber-optics and radio link systems. The DCX system follows standard network topology rules, so it’s important to observe cable and distance limitations and to use the proper network equipment, such as network bridges or routers, for extended distances. Information on basic network topology is covered in the appendix. The network shown in the above figure can be considered a small private network, since it contains a small number of nodes and it is not connected to any external networks. This system would require that three IP network addresses be assigned to the system components. The DCX control rack would need one address, and each client workstation would be assigned an address as well. If other client workstations were added to the system, each of them would also need a unique address. The hub device generally would not require an IP address. IP AddressingDevices connected to IP networks must be configured to talk to each other by giving each device an IP address and a subnet mask. The IP addresses consist of a 32-bit number that is divided into two parts: one part identifies the network on which a node resides, the other identified the node or host itself. The subnet mask is a 32-bit number that informs the PC about how to split the IP address into the node and network parts. The same subnet mask will be assigned to all the nodes in a simple network, while each node will have it’s own unique IP address. The following table lists an example IP address and mask written as binary numbers. It also shows the network and node addresses that are derived from the IP address and mask.
A network with three devices could be addressed as follows. Note that the subnet mask and network address are the same for all the devices, and the IP and node addresses are different for each device.
By now you’re wondering why all these numbers are needed. Since there are a lot of interconnected networks in the world, some means had to be developed to allow each computer to have it’s own address and avoid a situation where two computers are assigned the same address. Splitting the IP address in two parts, where one part identifies the particular network, and the other part identifies a particular device on that network was chosen. Consider the following address. Notice that the node address is the same as device 1 above, while the network address is different. This identifies that this device in on a different network than device 1.
Dotted Decimal NotationThe 32-bit binary addresses shown in the examples are difficult to work with, so a method of writing IP addresses called dotted-decimal notation was developed. The 32-bit IP address is divided into four 8-bit fields, called octets; the value of each field is written as a decimal number with the fields separated by periods.
IP Addressing RulesThe following guidelines should be followed when assigning IP addresses: · The bits used to define the node portion of an IP address can not be all “1” or all “0” bits. Node addresses consisting of all “1” or all “0” bits are considered subnet broadcast addresses. · The bits used to define the network portion of the IP address can not be all “0” bits. · The address 127.0.0.1 is reserved for the “loopback” function. · A network address is not allowed that has the four highest-order bits set to 1111. Address AssignmentIP addresses for networks that connect to the Internet are assigned by the Internet Assigned Numbers Authority or IANA, and are usually allocated in blocks to Internet Service Providers. Networks that don’t connect to the Internet can self-assign any address, but to avoid difficulty if you later decide to connect to the Internet, a special range of addresses is available for use on private networks. These private addresses may be self-assigned and are guaranteed not to conflict with addresses that are given out by the Internet address authority. Private, means these addresses are for use only on networks that don’t connect to the Internet. These addresses are specially programmed to be rejected by the Internets access routers, so if a private network accidentally becomes connected to the Internet, the routers will ignore it to prevent any conflicts. Should one of these private networks want to become connected to the public Internet at a later date, the user has two options. He can re-number his network with legal Internet addresses, or he can use an address translating router (NAT) or proxy to transform the internal private addresses to external public Internet addresses. The addresses that are reserved for private networks are: 10.0.0.0 – 10.255.255.255 (10/ 8
prefix) 172.16.0.0 – 172.31.255.255 (172.16/ 12
prefix) 192.168.0.0 – 192.168.255.255 (192.168/
16) prefix) The above address ranges are specified in the Internet document RFC 1918, “Address Allocation for Private Internets”. This document is available on the world wide web at http://ds.internic.net/rfc/rfc1918.txt . In deciding what address range to choose the following factors should be considered: · The RFC 1918 recommends using the 10.X.X.X address range first. · Microsoft uses the 10.X.X.X address range, and assigns itself 10.0.0.2 by default on Microsoft Small Business Server, which contains an address translating proxy. SBS can be used as a method to connect a private network to the Internet. · Many Internet Service providers use the 10.X.X.X address range to connect to clients on ISDN NAT routers. They usually reserve 10.0.0.1 for the client side port of the ISDN router. A NAT router is a device that can be used to connect a private network to the Internet. · If the 10.X.X.X range is used, a subnet mask should be used to allow for future router expansion. · Network support systems such as radio links, and mobile gear like laptop computers need to be considered. · A range of numbers for Remote Access Service (RAS) dial-in ports should be reserved. This makes it easy to add dial-in support to a network at a later date. Address Layout for a Private NetworkConsidering the factors listed above, the following address scheme can be used for a private DCX network. The 10.X.X.X range is chosen with a subnet mask of 255.255.255.0 . This will yield an expandable address space of 10.0.0.1 – 10.0.0.254 (10.0.0.0 and 10.0.0.255 are automatically assigned as broadcast addresses by the TCP/IP protocol rules). · The first 9 addresses 10.0.0.1 – 10.0.0.9 are reserved for network equipment, with 10.0.0.1 and 10.0.0.2 reserved for router and proxies devices. · Addresses 10.0.0.10 – 10.0.0.39 are used for fixed (non portable) PC’s and other fixed equipment. For example the first client PC would be 10.0.0.10, the second 10.0.0.11, etc. This will simplify setup and support. · Addresses 10.0.0.40 – 10.0.0.49 would be used for mobile equipment such as laptop computers that may be removed from the facility and used at another location. · Addresses 10.0.0.50 – 10.0.0.89 would be used for mobile equipment used by visitors or support personnel. · Addresses 10.0.0.90 – 10.0.0.99 are reserved for dial-in access expansion. · Reserve 10.0.0.100 – 10.0.0.254 for DCX servers. The serial number could be added to 100 to get the address: i.e. serial number 5 would be address 10.0.0.105 . Alternatively all the DCX units could be assigned the same address at all locations, such as 10.0.0.254. This would simplify setting up the clients.
|
