Telecom networks are undergoing substantial change--both in terms of technology as well as in the services they support. Ethernet displaced TDM transport, and IP is the underlying service delivery protocol for IPTV, VoIP, multimedia services, and of course, the Internet. These packet-based technologies have placed new demands on network-wide timing. This article will discuss how these changes are creating new requirements for accuracy, security, and availability of Network Time Protocol (NTP) based time, and then present what constitutes a true carrier-calls implementation for NTP synchronization that is needed to support these changes
NTP is an Internet protocol for synchronizing the clocks of computers and other equipment to a common time reference over a network. It is also a program that implements the protocol and controls the computer clocks. As originally conceived, NTP was designed to provide time to computer hosts, however, its use has grown and today it is the universally accepted method to synchronize system clocks in computers, servers, data communication equipment, and now, full service networks.
Historically, NTP has been deployed in the Internet, and then in telecom IT departments or datacenters for post-processing functions to support operational activities such as billing, AAA (access, authentication and accounting), and event log generation. Today, many uses for NTP have emerged, and it now is a critical technology that underlies carrier networks, services and operations.
Internet Protocol TV (IPTV) is the end-to-end delivery of video and related content using Internet Protocol over a managed network to consumers (i.e., not over the Internet). NTP-based time is used for many critical processes in IPTV service delivery. The most fundamental is synchronization of the IPTV headend and the set top box at the end points of the service delivery path. If their clocks are not aligned so that their data rates match, the set top box buffer will experience data overflow or data underflow. This result will lead to skipped frames and momentary freezing of skipping in the TV picture that are noticeable by the end user.
Two other examples are Conditional Access (CA) and QoS. To support these functions, the IPTV service infrastructure ó and therefore the time synchronization that supports it ó is distributed across a hierarchy of locations that includes the video headend office, the video service offices, the access network, and the customer premises.
Conditional Access licenses provide the basis for content usage and associated revenue generation. They are key to fulfilling digital rights management (DRM) agreements with the content owners as they prevent misuse and piracy that can steal revenue and lead to agreement violations.
Figure 1. Evolution to a highly distributed service infrastructure
Time on the systems that enforce CA licenses needs to be accurate and secure, for example, to ensure that operators can only show content during specific periods (such as after a DVDís release date). CA license keys also need secure time to prevent theft of service and alteration of the license usage duration. High capacity carrier grade NTP servers are required to handle the large transaction volumes generated by potentially millions of IPTV subscribers.
Quality of Service measurement is another key function in end-to-end IPTV service delivery. Some unique aspects of QoS measurement for IPTV are the number of measurement points in the network, the need for real-time monitoring, and the collection of metrics inside the customer premise at both the residential gateway and the set top box. Accurate and distributed time references are needed to enable QoS measurements across hops in the network, or between
Most time references that need to be synchronized are those serving customer endpoints. Providing NTP at all of these points calls for very high NTP server transaction capacity, bandwidth overhead, security, authentication, consistency, and network delay.
The IP Multimedia Subsystem (IMS) architecture has a multitude of functions that are distributed across many servers in the data center hosting the HSS/OSS/BSS functions. Billing and logging is for example decomposed into multiple functions that interface with many other servers. So what was previously one integrated software system on a classical mobile switching center (MSC) or public switched telephone network (PSTN) switch is now embodied in many servers with much richer and more complex processes and many potential interfaces. Time differences between logging and billing instances are widely distributed over a number of servers in a data center and also geographically distributed over multiple data centers.
The Call Detail Record (CDR) / Internet Protocol Detail Record (IPDR) is the basis for the billing contract with subscribers, for inter-carrier reconciliation, for taxation and for revenue statements that describe average revenue per user (ARPU). The IPDR is collected from a multitude of network instances depending on the specified service layer and application with well defined interfaces acting as collection points. Best effort NTP is no longer good enough for such a rich and complex service and application environment.
Fully redundant carrier grade NTP can deliver inherently precise and accurate timestamps with extremely stable performance.
The femtocell is a consumer device that connects to the service providerís network via residential broadband to provide UMTS-WCDMA services in residential areas. Frequency accuracy requirements for femtocells are typically in the range of 250 parts per billion (ppb).
Carrier grade NTP provides the necessary capacity and redundancy to serve thousands of femtocell devices on a single server and ensure high availability.
Closer to the edge
The pressure to meet all these needs, support new services and devices, improve network performance, accelerate fault diagnosis and recovery, and streamline billing, is driving the requirement for carrier grade NTP services. Operators must move accurate time closer to the network edge to manage end-user devices.
NTP has proven its value to both enterprise and carrier operators. Telecommunications network and service transitions to packet-based technologies increases the importance of deploying a carrier grade NTP implementation.
About the Author
Barry Dropping is currently the Director of Product Line Management for Symmetricomís communications solutions. In this position, Mr. Dropping directs the development of next generation synchronization products for fixed, mobile and cable operators. Mr. Dropping joined Symmetricom in October 1999 following its acquisition of Hewlett-Packardís communications synchronization business. Mr. Dropping spent 15 years at Hewlett-Packard where he served in a variety of management positions across engineering, marketing and operations. Mr. Dropping holds a BS in Electrical Engineering Technology from the DeVry Institute.