Module Overview
When migrating to a Voice over IP (VoIP) network, all network requirements, including power and capacity planning, must be examined. In addition, congestion avoidance techniques should be implemented. This module highlights the basic issues and defines the initial steps to take to ensure a functional VoIP implementation.
7.1 Planning for Implementation of Voice in a Campus
7.1.1 Converged Network Benefits
The benefits of packet telephony versus circuit-switched telephony are as follows:
More efficient use of bandwidth and equipment: Traditional telephony networks use a 64-kbps channel for every voice call. Packet telephony shares bandwidth among multiple logical connections and offloads traffic volume from existing voice switches.
Lower costs for telephony network transmission: A substantial amount of equipment is needed to combine 64-kbps channels into high-speed links for transport across the network. Packet telephony statistically multiplexes voice traffic alongside data traffic. This consolidation represents substantial savings on capital equipment and operations costs.
Consolidated voice and data network expenses: Data networks that function as separate networks to voice networks become major traffic carriers. The underlying voice networks are converted to utilize the packet-switched architecture to create a single integrated communications network with a common switching and transmission system. The benefit is significant cost savings on network equipment and operations.
Increased revenues from new services: Packet telephony enables new integrated services, such as broadcast-quality audio, unified messaging, and real-time voice and data collaboration. These services increase employee productivity and profit margins well above those of basic voice services. In addition, these services enable companies and service providers to differentiate themselves and improve their market position.
Greater innovation in services: Unified communications use the IP infrastructure to consolidate communication methods that were previously independent; for example, fax, voice mail, e-mail, wireline telephones, cellular telephones, and the Web. The IP infrastructure provides users with a common method to access messages and initiate real-time communications—independent of time, location, or device.
Access to new communications devices: Packet technology can reach devices that are largely inaccessible to the time-division multiplexing (TDM) infrastructures of today. Examples of such devices are computers, wireless devices, household appliances, personal digital assistants, and cable set-top boxes. Intelligent access to such devices enables companies and service providers to increase the volume of communications they deliver, the breadth of services they offer, and the number of subscribers they serve. Packet technology, therefore, enables companies to market new devices, including videophones, multimedia terminals, and advanced IP phones.
Flexible new pricing structures: Companies and service providers with packet-switched networks can transform their service and pricing models. Because network bandwidth can be dynamically allocated, network usage no longer needs to be measured in minutes or distance. Dynamic allocation gives service providers the flexibility to meet the needs of their customers in ways that bring them the greatest benefits.
7.1.2 VoIP Network Components
The basic components of a VoIP network are:
IP phones: Provide IP voice to the desktop.
Gatekeeper: Provides connection admission control (CAC), bandwidth control and management, and address translation.
Gateway: Provides translation between VoIP and non-VoIP networks, such as the public switched telephone network (PSTN). It also provides physical access for local analog and digital voice devices, such as telephones, fax machines, key sets, and PBXs.
Multipoint control unit (MCU): Provides real-time connectivity for participants in multiple locations to attend the same videoconference or meeting.
Call agent: Provides call control for IP phones, CAC, bandwidth control and management, and address translation.
Application servers: Provide services such as voice mail, unified messaging, and Cisco CallManager Attendant Console.
Videoconference station: Provides access for end-user participation in videoconferencing. The videoconference station contains a video capture device for video input and a microphone for audio input. The user can view video streams and hear the audio that originates at a remote user station.
Other components, such as software voice applications, interactive voice response (IVR) systems, and soft phones, provide additional services to meet the needs of enterprise sites.
7.1.3 Traffic Characteristics of Voice and Data
Voice traffic has extremely stringent quality of service (QoS) requirements. Voice traffic usually generates a smooth demand on bandwidth and has minimal impact on other traffic as long as voice traffic is managed.
Although voice packets are typically small (60 to 120 bytes), they cannot tolerate delay or drops. The result of delays and drops is often unacceptable voice quality. Because drops cannot be tolerated, User Datagram Protocol (UDP) is used to package voice packets. TCP retransmit capabilities have no value.
For voice quality, the delay should be no more than 150 ms (one-way requirement) and less than 1 percent packet loss.
A typical voice call requires 17 to 106 kbps of guaranteed priority bandwidth, plus an additional 150 bps per call for voice-control traffic. Multiplying these bandwidth requirements by the maximum number of calls expected during the busiest time period indicates the overall bandwidth required for voice traffic.
The QoS requirements for data traffic vary greatly.
Different applications (for example, a human resources application versus an automated teller machine [ATM] application) may make greatly different demands on the network. Even different versions of the same application may have varying network traffic characteristics.
Data traffic can demonstrate either smooth or bursty characteristics, and it differs from voice and video in terms of delay and drop sensitivity. Almost all data applications can tolerate some delay and generally can tolerate high drop rates.
Because data traffic can tolerate drops, the retransmit capabilities of TCP become important and, as a result, many data applications use TCP.
It is important to be able to identify different types of traffic that move over networks. With TCP/IP, most applications can be identified by their use of TCP or UDP port numbers, and with TCP, a stream of traffic usually occurs.
However, some applications use dynamic port numbers that make classifications more difficult. Cisco IOS software supports network-based application recognition (NBAR), which can be used to recognize dynamic port applications.
7.1.4 VoIP Call Flow
VoIP calls can contend with normal client data for bandwidth. If both the client PC and the VoIP phone are on the same VLAN, each will try to use the available bandwidth without consideration of the other device. To avoid this issue, use two VLANs to allow separation of VoIP and client data. After data is separated, QoS can be applied to prioritize the VoIP traffic as it traverses the network.
A major component of designing a successful IP telephony network is properly provisioning the network bandwidth. You can calculate the required bandwidth by adding the bandwidth requirements for each major application, including voice, video, and data. This sum represents the minimum bandwidth requirement for any given link, and it should not exceed approximately 75 percent of the total available bandwidth for the link.
From a traffic standpoint, an IP telephony call consists of two traffic types, as illustrated in Figure 
using a Cisco CallManager:
Voice carrier stream: Real-Time Transport Protocol (RTP) packets that contain the actual voice samples.
Call control signaling: Packets belonging to one of several protocols—those used to set up, maintain, tear down, or redirect a call, depending upon call endpoints. Examples are H.323 or Media Gateway Control Protocol (MGCP).
A VoIP packet consists of the voice payload, RTP header, UDP header, IP header, and Layer 2 encapsulation. The IP header is 20 bytes, the UDP header is 8 bytes, and the RTP header is 12 bytes. The link layer overhead varies in size according to the Layer 2 media used; Ethernet requires 18 bytes of overhead. The voice payload size and the packetization period are device dependent.
Coder-Decoders (codecs) convert the analog voice to a digital signal format. This technology has been used for years to convert a telephone signal into a 64,000 bps digital signal (DS0) for use on TDM-based systems. Today, an IP phone uses a G.711 codec for normal voice digitization. G.711 is the only type supported for the Cisco Conference Connection and Personal Assistant applications. G.729 is another supported codec that provides compression of the voice traffic down to 8 kbps. Cisco VoIP equipment supports G.711 and G.729, along with several other common industry standards.
7.1.5 Auxiliary VLANs
Some Cisco Catalyst switches offer a unique feature called an “auxiliary VLAN” or a “voice VLAN.” Auxiliary VLANs allow you to overlay a voice topology onto a data network. You can segment phones into separate logical networks, even though the data and voice infrastructure are physically the same.
Auxiliary VLANs place the phones into their own VLANs without any end-user intervention. Furthermore, these VLAN assignments can be seamlessly maintained, even if the phone is moved to a new location. The user simply plugs the phone into the switch, and the switch provides the phone with the necessary VLAN information. By placing phones into their own VLANs, network administrators gain the advantages of network segmentation and control. Furthermore, network administrators can preserve their existing IP topology for the data end stations. IP phones can be easily assigned to different IP subnets using standards-based DHCP operation.
With the phones in their own IP subnets and VLANs, network administrators can more easily identify and troubleshoot network problems. Additionally, network administrators can create and enforce QoS or security policies. Auxiliary VLANs enable Cisco network administrators to gain all the advantages of physical infrastructure convergence while maintaining separate logical topologies for voice and data terminals. This creates the most effective way to manage a multiservice network.
7.1.6 QoS
Almost any network can take advantage of QoS for optimum efficiency, whether it is a small corporate network, an Internet service provider (ISP), or an enterprise network. QoS utilizes features and functionality to meet the networking requirements of applications sensitive to loss, delay, and delay variation (jitter). QoS allows preference to be given to critical application flows for the available bandwidth.
The Cisco IOS implementation of QoS software provides these benefits:
Priority access to resources: Administrators can control which traffic is allowed to access specific network resources, such as bandwidth, equipment, and WAN links. Critical traffic can take possession of a resource because the QoS implementation drops low-priority frames.
Efficient management of network resources: If network management and accounting tools indicate that specific traffic is experiencing latency, jitter, or packet loss, you can use QoS tools to adjust how that traffic is handled.
Tailored services: ISPs can offer carefully tailored grades of service to their customers. For example, an ISP can offer one service level agreement (SLA) to a customer website that receives 3,000 to 4,000 hits per day and another to a site that receives only 200 to 300 hits per day.
Coexistence of mission-critical applications: Mission-critical business applications receive priority access to network resources while providing adequate processing for applications that are not delay sensitive. Multimedia and voice applications tolerate little latency and require priority access to resources. Other delay-tolerant traffic traversing the same link, such as Simple Mail Transfer Protocol (SMTP) over TCP, can still be adequately serviced.
7.1.7 Importance of High Availability for VoIP
The traditional telephony network strives to provide 99.999 percent uptime to the user. This corresponds to 5.25 minutes per year of downtime. Many data networks cannot make the same claim. To provide telephony users the same, or close to the same, level of service as they experience with traditional telephony, the reliability and availability of the data network takes on new importance.
Reliability is a measure of how resilient a network can be. Efforts to ensure reliability include choosing hardware and software with a low mean time between failure, or installing redundant hardware and links. Availability is a measure of how accessible the network is to the users. When a user wants to make a call, for example, the network should be accessible to that user. Efforts to ensure availability include installing proactive network management to predict failures before they happen, and taking steps to correct problems in the design of the network as it grows.
When the data network goes down, it may not come back up for minutes or even hours. This delay is unacceptable for telephony users. Local users with network equipment, such as voice-enabled routers, gateways, or switches for IP phones, now find that their connectivity is terminated. Administrators must provide an uninterruptible power supply (UPS) to these devices in addition to providing network availability. Previously, users received their power directly from the telephone company central office or through a UPS that was connected to a keyswitch or PBX in the event of a power outage. Today, the network devices must continue to function, provide service to the end devices, and possibly (as with Power over Ethernet [PoE]) supply power to end devices.
Note:
Cisco has the option of using DC power with many of its routers, which allows power to be distributed from a “battery bank” that is continuously being charged. When a power outage occurs, the batteries supply DC to the equipment. Battery banks are very common in the telephone industry.
Network reliability comes from incorporating redundancy into the network design. In traditional telephony, switches have multiple redundant connections to other switches. If either a link or a switch becomes unavailable, the telephone company can easily re-route calls. This is why telephone companies can claim a high availability rate.
High availability encompasses many areas of the network. In a fully redundant network, the following components need to be duplicated:
Servers and call managers
Access layer devices, such as LAN switches
Distribution layer devices, such as routers or multilayer switches
Core layer devices, such as multilayer switches
Interconnections, such as WAN links and PSTN gateways, even through different providers
Power supplies and UPSs
In some data networks, a high level of availability and reliability is not critical enough to warrant financing the hardware and links required to provide complete redundancy. But if voice is layered onto the network, these requirements need to be revisited.
With Cisco Architecture for Voice, Video and Integrated Data (AVVID) technology, Cisco CallManager clusters provide a way to design redundant hardware. When using gatekeepers, you can configure backup devices as secondary gatekeepers in case the primary gatekeeper fails. Redundant devices and Cisco IOS services, like Hot Standby Router Protocol (HSRP), also provide high availability. For proactive network monitoring and trouble reporting, a network management platform such as CiscoWorks2000 provides a high degree of responsiveness to network issues.
7.1.8 Power Requirements in Support of VoIP
Accurate calculations of power requirements are critical for an effective IP telephony solution. IP phones are best implemented with PoE. Power can be supplied to the IP phones directly from Cisco Catalyst switches with inline power capabilities or by inserting a Cisco Catalyst Inline Power Patch Panel. In addition to IP phones, failover power and total load must be considered for all devices in the IP telephony availability definition, including Building Distribution and Campus Backbone submodules, gateways, Cisco CallManager, and other servers and devices. Power calculations must be network-based rather than device-based. Also, as with wireless access points, VoIP phones are best implemented with Power over Ethernet (PoE).
To provide highly available power protection, you need either a UPS with a minimum battery life of 1 hour for power system failures, or a generator. This solution must include UPS or generator backup for all devices associated with the IP telephony network. In addition, consider UPS systems that have auto-restart capability and a service contract for 4-hour support response.
Recommendations for IP telephony high-availability power and environment include the following:
UPS and generator backup
UPS systems with auto-restart capability
UPS system monitoring
4-hour service response contract for UPS system problems
Recommended equipment operating temperatures maintained at all times
No hay comentarios:
Publicar un comentario