www.networkbulls.com
Best Institute for CCNA CCNP CCSP CCIP CCIE Training in India
M-44, Old Dlf, Sector-14 Gurgaon, Haryana, India
Call: +91-9654672192
WAN technologies provide network connectivity for the Enterprise Edge and remote branch edge locations as well as the Internet. Many WAN choices are available, and new ones are continually emerging. When you're selecting WAN transport technologies, it is important to consider factors such as cost, speed, reliability, hardware, and media. In addition, enterprise branch offices can take advantage of cable and DSL technologies for remote connectivity back to the headquarters or main office.
WAN Defined
Wide-area networks (WANs) are communication networks that can span great distances to provide connectivity. They generally are offered by service providers or carriers. WANs typically carry data traffic, but many now support voice and video as well. Service providers charge fees for providing WAN services or communications to their customers. Sometimes the term "service" is referred to as the WAN communications provided by the carrier.
Figure 5-1 depicts the Enterprise Edge with campus backbone, Internet, and MPLS clouds.
Figure 5-1. Enterprise WAN
When designing a WAN, you should become familiar with the design's requirements, which are typically derived from these two important goals:
WAN Connection Modules
The Enterprise Edge can have multiple WAN interconnections. Common connectivity modules include but are not limited to the Internet, the demilitarized zone (DMZ), and the WAN. Internet service providers (ISPs) offer many connectivity options for the Internet and DMZ modules of the Enterprise Edge. Internal WAN connectivity between an organization's headquarters and remote sites generally is across a service provider or carrier network. PSTN connectivity still exists for teleworkers and more recently because of the increasing use of VoIP offnet services.
WAN technologies such as Frame Relay exist for point-to-point (P2P) and multipoint WAN services. Service providers also offer full IP WAN solutions such as MPLS where the Enterprise Edge router interacts with service providers at Layer 3. Public WAN connections over the Internet are also available through the use of cable and/or DSL technologies. Typically, these services do not provide any guarantee of network availability, as do Frame Relay and MPLS network solutions.
Figure 5-2 illustrates the use of modules, or blocks, in the Enterprise Edge.
Figure 5-2. WAN Interconnections
WAN Comparison
Table 5-2 examines some WAN technologies and highlights some common factors that are used to make WAN technology selections. This information also reflects the different characteristics of each WAN technology. However, keep in mind that your service provider offerings limit the WAN technology choices available to you during your selection.
Dialup
Dialup technology provides connectivity over the PSTN using analog modems. Although the bandwidth is relatively low, the availability of analog is very widespread. Dialup connectivity is ideal for low-bandwidth conversations of 56 kbps or less. Despite the high availability of dialup technology over analog lines, it is generally not a viable option anymore. However, a common use of dialup is when a remote worker or teleworker uses it as a backup network solution if his or her DSL or cable connection goes down.
ISDN
Integrated Services Digital Network (ISDN) is an all-digital phone line connection that was standardized in the early 1980s. ISDN allows both voice and data to be transmitted over the digital phone line instead of the analog signals used in dialup connections. ISDN provides greater bandwidth and lower latency compared to dialup analog technology. ISDN comes in two service types—Basic Rate Interface (BRI) and Primary Rate Interface (PRI).
ISDN is comprised of digital devices and reference points. ISDN devices consist of terminals, terminal adapters, network-termination, line-termination, and exchange-termination equipment. Native ISDN devices are referred to as terminal equipment 1 (TE1), and nonnative ISDN is referred to as terminal equipment 2 (TE2). However, TE2-type devices can be connected to an ISDN system with the help of a terminal adapter (TA).
Working toward the service provider after the TE1 and TE2 devices, the next connection devices are network termination 2 (NT2) and network termination 1 (NT1). These connection devices connect the five-wire to the two-wire local loop. In North America, the NT1 is a CPE device or customer premises equipment. This means that the customer, not the carrier, provides the device. However, in most other parts of the world, the carrier provides the NT1.
Note
ISDN has a series of reference points that define logical interfaces between ISDN devices such as TAs and NT1s:
- T— Reference point between NT2 and NT1 devices
- U— Reference point between NT1 and the line termination equipment in the carrier's network
Figure 5-3 illustrates how ISDN devices and reference points relate to each other.
Figure 5-3. ISDN Devices and Reference Points
ISDN BRI Service
ISDN BRI consists of two B channels and one D channel (2B+D). Both of the BRI B channels operate at 64 kbps and carry user data. The D channel handles the signaling and control information and operates at 16 kbps. Another 48 kbps is used for framing control and other overhead, for a total bit rate of 192 kbps.
ISDN PRI Service
ISDN PRI service offers 23 B channels and one D channel (23B+D) in both North America and Japan. Each channel (including the D channel) operates at 64 kbps, for a total bit rate of 1.544 Mbps, including overhead. In other parts of the world, such as Europe and Australia, the ISDN PRI service provides 30 B channels and one 64-kbps D channel.
Frame Relay
Frame Relay is a connection-oriented Layer 2 WAN protocol. It is similar to X.25 but has faster performance due to the lack of error checking and retransmitting features. The data link layer establishes connections in Frame Relay using a DTE device such as a router and a DCE device such as a frame switch.
In the early 1980s, networks were growing using more and more point-to-point leased-line connections. Full-mesh connections were used to ensure redundancy between sites. However, full-mesh network configurations presented a larger cost because of the number of leased lines needed. A full mesh requires that each site have a connection to the other sites participating in the full mesh. For example, if you have five WAN sites, each site needs four leased lines to the other sites to complete the full mesh. However, because Frame Relay uses a cloud of switches, each site can use only one connection to the Frame Relay cloud and then can be configured to emulate a full mesh. This reduces the leased-line cost and requires only one leased line per site. This allows the network to achieve full-mesh-like behavior needed for network redundancy.
Frame Relay circuits between sites can be either permanent virtual circuits (PVC) or switched virtual circuits (SVC). PVCs are used more predominantly due to the connections' permanent nature. SVCs, on the other hand, are temporary connections created for each data transfer session.
A point-to-point PVC between two routers or endpoints uses a data-link connection identifier (DLCI) to identify the local end of the PVC. The DLCI is a locally significant numeric value that can be reused through the Frame Relay WAN if necessary.
Local Management Interface
Frame Relay uses a signaling protocol between the Frame Relay router and the Frame Relay switch called the Local Management Interface (LMI). The LMI protocol sends periodic keepalive messages and notifications of additions or removals of PVCs. Three types of LMI protocols are available. The service provider usually informs you on which one to use. LMI also offers a number of features or extensions, including global addressing, virtual circuit status messages, and multicasting. By default, Cisco routers will try all three LMI types until a match is found.
Discard Eligibility
The Discard Eligibility (DE) bit is used in Frame Relay to identify whether a frame has lower importance than other frames. The DE bit is part of the Frame Relay header and can have a value of 1 or 0. Routers or DTE devices can set the value of the DE bit to 1 to indicate that the frame has lower importance than frames marked with a 0. During periods of congestion, the Frame Relay network discards frames marked with the DE bit of 1 before those marked with 0. This reduces the chance of critical data being dropped, because you can identify what traffic gets marked with a DE bit value of 1.
Time-Division Multiplexing
Time-Division Multiplexing (TDM) is a type of digital multiplexing in which multiple channels such as data, voice, and video are combined over one communication medium by interleaving pulses representing bits from different channels. Basic DS0 channel bandwidth is defined at 64 kbps. In North America, a DS1 or T1 circuit provides 1.544 Mbps of bandwidth consisting of 24 time slots of 64 kbps each and an 8-kbps channel for control information. In addition, a DS3 or T3 circuit provides 44.736 Mbps of bandwidth. Other parts of the world, such as Europe, follow E1 standards, which allow for 30 channels at 2.048 Mbps of bandwidth. Service providers can guarantee or reserve the bandwidth used on TDM networks. The customers' TDM transmissions are charged for their exclusive access to these circuits. On the other hand, packet-switched networks typically are shared, thereby allowing the service providers more flexibility in managing their networks and the services they offer.
SONET/SDH
The architecture of Synchronous Optical Network/Synchronous Digital Hierarchy (SONET/SDH) is circuit-based and delivers high-speed services over an optical network. The term SONET is defined by the American National Standards Institute (ANSI) specification, and SDH is defined by the International Telecommunication Union (ITU). SONET/SDH guarantees bandwidth and has line rates of 155 Mbps to more than 10 Gbps. Common circuit sizes are OC-3, or 155 Mbps, and OC-12, or 622 Mbps.
SONET/SDH uses a ring topology by connecting sites and providing automatic recovery capabilities and has self-healing mechanisms. SONET/SDH rings support ATM or packet over SONET (POS) IP encapsulations. The Optical Carrier (OC) rates are the digital bandwidth hierarchies that are part of the SONET/SDH standards. The optical carrier speeds supported are as follows:
- OC-1 = 51.85 Mbps
- OC-3 = 155.52 Mbps
- OC-12 = 622.08 Mbps
- OC-24 = 1.244 Gbps
- OC-48 = 2.488 Gbps
- OC-192 = 9.952 Gbps
- OC-255 = 13.21 Gbps
Figure 5-4 shows an OC-48 SONET ring with connections to three sites that share the ring.
Figure 5-4. SONET/SDH
Multiprotocol Label Switching
MPLS is technology for the delivery of IP services using labels (numbers) to forward packets. In normal routed environments, packets are forwarded by the router performing a Layer 3 destination address lookup and rewriting the Layer 2 addresses. MPLS functions by encapsulating packets with headers that include the label information. As soon as packets are marked with a label, specific paths through the network can be designed to correspond to that distinct label. MPLS labels can be based on parameters such as source addresses, Layer 2 circuit ID, or QoS value. The labels can be used to implement traffic engineering by overriding the routing tables. MPLS packets can run over most Layer 2 technologies, such as ATM, Frame Relay, POS, and Ethernet. The goal of MPLS is to maximize switching using labels and minimize Layer 3 routing.
In most MPLS implementations, the equipment is called the customer edge (CE) and provider edge (PE) routers. Typically the customer-owned internal WAN router peers with the CE router. The CE router then connects to the PE router, which is the ingress to the MPLS service provider network. The PE router is in the service provider network.
Figure 5-5 shows an MPLS WAN and how the CE routers connect to the provider.
Figure 5-5. MPLS
Other WAN Technologies
This section briefly discusses other WAN technologies that are becoming very popular in the network access space as well as some other service provider architectures.
Digital Subscriber Line
Digital Subscriber Line (DSL) is a technology that provides high-speed Internet data services over ordinary copper telephone lines. It achieves this by using frequencies that are not used in normal voice telephone calls.
The term xDSL describes the various competing forms of DSL available today. Some of the DSL technologies available include asymmetric (ADSL), symmetric (SDSL), high bit rate (HDSL), very high bit rate (VDSL), rate-adaptive (RADSL) and IDSL (based on ISDN).
Table 5-3 summarizes the types of DSL specifications.
| Service | Maximum Distance to Central Office | Maximum Upload Speed | Maximum Download Speed | Notes |
|---|---|---|---|---|
| Full-rate ADSL | 18,000 ft (5500 m) | 1500 kbps | 9 Mbps | Asymmetrical. |
| ADSL G.lite | 18,000 ft (5500 m) | 384 kbps | 1.5 Mbps | No splitter is required. |
| RADSL | 18,000 ft (5500 m) | 384 kbps | 8 Mbps | Rate adapts based on distance and quality. |
| IDSL | 35,000 ft (10,070 m) | 144 kbps | 144 kbps | DSL over ISDN (BRI). |
| SDSL | 22,000 ft (6700 m) | 2.3 Mbps | 2.3 Mbps | Targets T1 replacement. Symmetrical DSL service. |
| HDSL | 18,000 ft (5500 m) | 1.54 Mbps | 1.54 Mbps | Four-wire, similar to T1 service. |
| HDSL-2 | 24,000 ft (7333 m) | 2 Mbps | 2 Mbps | Two-wire version of HDSL or four-wire at 2 times the rate. |
| VDSL | 3000 ft (916 m) | 16 Mbps | 52 Mbps | Few installations. |
ADSL is the most popular DSL technology and is widely available. The key to ADSL is that the downstream bandwidth is asymmetric or higher than the upstream bandwidth. Some limitations include that ADSL can be used only in close proximity to the local DSLAM, typically less than 2 km. The local DSLAM, or digital subscriber line access multiplexer, allows telephone lines to make DSL connections to the Internet. Download speeds usually range from 768 kbps to 9 Mbps, and upload speeds range from 64 kbps to 1.5 Mbps. Generally, the equipment used is a DSL modem or (CPE) router that connects back to the ISP's DSLAM.
Although DSL is primarily used in the residential community, this technology can also be used as a WAN technology for an organization. However, keep in mind that because this is a public network connection over the Internet, it is recommended that this technology be used in conjunction with a firewall/VPN solution back into your corporate enterprise network. The high speeds and relatively low cost make this a very popular Internet access WAN technology.
Cable
Broadband cable is a technology used to transport data using a coaxial cable medium over cable distribution systems. The equipment used on the remote-access side is the cable modem, which connects to the Cable Modem Termination System (CMTS) on the ISP side. The Universal Broadband Router (uBR) provides the CMTS services and is deployed at the cable company headend. The uBR forwards traffic upstream through the provider's WAN core or the local PSTN, depending on the services being provided.
The Data Over Cable Service Interface Specifications (DOCSIS) protocol defines the cable procedures that the equipment needs to support. DOCSIS 2.0 was released in 2002 and remains the current version that most cable modems use today. DOCSIS 3.0 specifications released in 2006 include support for IPv6 and channel bonding.
Figure 5-6 illustrates how a cable modem connects to the CMTS.
Figure 5-6. Data Over Cable
Wireless
Wireless as a technology uses electromagnetic waves to carry the signal between endpoints. Everyday examples of wireless technology include cell phones, wireless LANs, cordless computer equipment, and satellite television.
- Mobile wireless— Consists of cellular applications and mobile phones. Most wireless technologies such as the second and third generations are migrating to more digital services to take advantage of the higher speeds. Mobile wireless technologies include GSM, GPRS, and UMTS:
- - GSM— Global system for mobile communications. A digital mobile radio standard that uses Time-Division Multiplex Access (TDMA) technology in three bands—900, 1800, and 1900 MHz. The data transfer rate is 9600 bps and includes the ability to roam internationally.
- - GPRS— General Packet Radio Service. Extends the capability of GSM speeds from 64 kbps to 128 kbps.
- - UMTS— Universal Mobile Telecommunications Service. Also known as 3G broadband. Provides packet-based transmission of digitized voice, video, and data at rates up to 2.0 Mbps. UMTS also provides a set of services available to mobile users, location-independent throughout the world.
- Wireless LAN— WLANs have increased too in both residential and business environments to meet the demands of LAN connections over the air. Commonly called IEEE 802.11a/b/g or Wi-Fi wireless networks. Currently, 802.11n is in development and provides typical data rates of 200 Mb/s. The growing range of applications includes guest access, voice over wireless, and support services such as advanced security and location of wireless endpoints. A key advantage of WLANs is the ability to save time and money by avoiding costly physical layer wiring installations.
- Bridge wireless— Wireless bridges connect two separate wireless networks, typically located in two separate buildings. This technology enables high data rates for use with line-of-sight applications. When interconnecting hard-to-wire sites, temporary networks, or warehouses, a series of wireless bridges can be connected to provide connectivity.
Figure 5-7 shows bridge wireless and wireless LANs.
Figure 5-7. Wireless Implementations
Note
Additional information on wireless LANs is provided in Chapter 4, "Wireless LAN Design."
Dark Fiber
Dark fiber is fiber-optic cable that has been installed in the ground or where right-of-way issues are evident. To maintain signal integrity and jitter control over long distances, signal regenerators are used in some implementations. The framing for dark fiber is determined by the enterprise, not the provider. The edge devices can use the fiber just like within the enterprise, which allows for greater control of the services provided by the link. Dark fiber is owned by service providers in most cases and can be purchased similarly to leased-line circuits for use in both the MAN and WAN. The reliability of these types of links also needs to be designed by the enterprise and is not provided by the service provider. This contrasts with SONET/SDH, which has redundancy built into the architecture.
Dense Wave Division Multiplexing
Dense Wave Division Multiplexing (DWDM) increases fiber optic's bandwidth capabilities by using different wavelengths of light called channels over the same fiber strand. It maximizes the use of the installed base of fiber used by service providers and is a critical component of optical networks. DWDM allows for service providers to increase the services offered to customers by adding new bandwidth to existing channels on the same fiber. DWDM lets a variety of devices access the network, including IP routers, ATM switches, and SONET terminals.
Figure 5-8 illustrates the use of DWDM using Cisco ONS devices and a SONET/SDH ring.
Figure 5-8. DWDM
Ordering WAN Technology and Contracts
When you order WAN transport technology, early planning is key. It usually takes at least 60 days for the carrier to provision circuits. Generally, the higher a circuit's capacity, the more lead time is required to provision. When ordering bandwidth overseas, a lead time of 120 days is fairly common.
WAN transport in most cases includes an access circuit charge and, at times, distance-based charges. However, some carriers have eliminated TDM distance-based charges because T1s are readily available from most carriers. In rare cases, construction is necessary to provide fiber access, which requires more cost and time delays. You should compare pricing and available WAN technology options from competing carriers.
When ordering Frame Relay and ATM, a combination of access circuit charges, per-PVC charges, and per-bandwidth Committed Information Rate (CIR) charges are customary. CIR is the rate that the provider guarantees it will provide. Some carriers set the CIR to half the circuit's speed, thereby allowing customers to burst 2 times above the CIR. Frame Relay speeds can be provisioned up to T3 speeds, but typically they are less than 10 Mbps.
MPLS VPNs have been very competitive with ATM and Frame Relay rates. Service providers are offering MPLS VPNs with higher bandwidth at lower rates to persuade their customers away from traditional ATM and Frame Relay services. However, other service providers see more value in MPLS VPNs and price them higher than ATM and Frame Relay because of the added benefits of traffic engineering.
When you're selecting a standard carrier package, it takes about a month to contract a WAN circuit. If you want to negotiate a detailed service level agreement (SLA), expect to take another five months or more, including discussions with the service provider's legal department. The bigger the customer, the more influence it has over the SLAs and the contract negotiations.
Contract periods for most WAN services are one to five years. Contracts are usually not written for longer durations because of the new emerging technologies and better offerings from providers. An exception is dark fiber, which is usually contracted for a 20-year term. In this case you also want to have the right of nonreversion written in the SLA. This means that no matter what happens to the service provider, the fiber is yours for the 20-year period.
No comments:
Post a Comment