The Development of SDH and the Future Path of MSTP
Regarding the future direction of the SDH system, there are two voices in the industry: one holds that the SDH system needs continuous enhancement and improvement to secure its position as the foundation of the next-generation network architecture, a view mainly coming from traditional telecom operator groups; the other holds that the IP network architecture is the future of communications, and implementing a "cap and grow" strategy (i.e., ceasing all new TDM investment, with all upgrades, expansions, and new builds adopting next-generation technology products) to simplify the SDH network and ultimately abandon the SDH network architecture is the wise move.
Traditional SDH suffers from numerous problems, including bandwidth bottlenecks, overly complex provisioning in multi-layer network structures, and support for only a single type of service. Some operators in Europe, East Asia, and India have already completely abandoned the SDH technology system in new network builds (especially in metropolitan area networks). However, regardless of its ultimate fate, the foundational position of the SDH network will not change for at least the next few decades. The current SDH network has already grown so massive that traditional telecom operators cannot calmly and easily walk away from it (according to surveys, the total investment by various Chinese telecom operators in SDH is around RMB 200 billion). Moreover, although the ARPU values of SDH-based services are generally not high, they remain an important source of profit for these operators. After the pain of the network bubble, without sufficient foreseeable revenue to make their "cap and grow" plans worthwhile, operators will not blindly pursue new technologies. Instead, they will give more consideration to reducing capex and opex while maintaining smooth network evolution.
Born of the Times: MSTP Has a Promising Outlook
As market competition intensifies, traditional telecom operators must segment their customer markets and provide differentiated services, and network support capability and operation & maintenance capability are the keys to truly implementing the concept of differentiation. SDH systems have been gradually pushed to the edge of the network by increasingly mature WDM systems, and being at the network edge means diversity in access services (signals). Although SDH can transmit almost all data formats (IP, RF, ATM, etc.) through corresponding technical means such as mapping and concatenation, simply performing fixed encapsulation and transparent transmission of data and providing Layer 2 switching and local aggregation functions, the bandwidth of traditional SDH systems is provisioned through centralized network management systems and increases/decreases in multiples of 4, which contradicts the dynamic characteristics of data service bandwidth.
"One should take a long-term view of things." For traditional telecom operators facing rapidly growing data service demands, the urgent task is to seek a transmission solution based on the SDH network architecture that supports multiple services, offers high integration and high intelligence, and is standardized, so as to carry both TDM and data services simultaneously, dynamically configure channel bandwidth, improve and perfect the existing SDH network, integrate the separated SDH layer, ATM layer, and IP layer, protect existing investments, and enhance network survivability.
Thus, MSTP, known as the next-generation SDH, emerged at the right time. Strictly speaking, MSTP is not an entirely new technology; rather, it embeds or integrates mature existing technologies such as Ethernet, ATM, RPR, ESCON, FICON, Fibre Channel, and MPLS into SDH through means such as mapping, VC virtual concatenation, GFP, LCAS, and bus technology. As such, MSTP products are very much like a versatile box with extraordinary magic. The technical positioning of MSTP lies in integrating TDM and Ethernet Layer 2 switching, achieving intelligent control and management of data through Layer 2 switching, optimizing the transmission of data in SDH channels, and effectively solving the problems of single-service support and fixed bandwidth in ADM/DXC equipment, the high cost of ATM equipment, and the limited networking capability and QoS issues of IP equipment. At the early stage of MSTP's emergence, all investors, including chip manufacturers, equipment manufacturers, and operators, placed high hopes on it, hoping that MSTP would become as indispensable to the SDH network as routers are to the Internet.
In recent years, MSTP equipment has lived up to expectations in the global market, with sales climbing steadily and proving especially popular in the metropolitan area network field—indeed, "blooming as brilliantly as summer flowers." In North America, the Baby Bells, known for their conservatism, announced plans last year to deploy next-generation SDH products—MSTP—which fully demonstrates the formidable momentum of MSTP. Subsequently, Bell South and SBC also announced plans to deploy MSTP products. At SuperComm 2004, numerous vendors showcased their latest MSTP products. According to IDC forecasts, the global MSTP market will grow at an annual rate of 18%, reaching USD 3.2 billion by 2007. This "brilliant" achievement is closely related to the fact that MSTP technology and products are entirely driven by market demand—"the most suitable is the best."
Technical Maturity and Business Models: Two Hurdles MSTP Must Cross
Looking at the domestic market, the development of MSTP over the past few years has not been entirely satisfactory. Although both operators and equipment manufacturers are optimistic about the market prospects of MSTP, large-scale application of MSTP in China remains relatively rare.
One important reason is that the maturity of MSTP equipment is still insufficient. Over the past few years, MSTP technology has been in continuous development. From MADM in 2000, ATM switching and L2 switching in 2001, to RPR in 2002 and GFP/LCAS in 2003, and now MPLS, MSTP adds new content every year. Zhang Chengliang, Director of the Technology Department at China Telecom Beijing Research Institute, believes that MSTP has been in continuous evolution; although its functions have been steadily enhanced, the technology is not yet mature. Currently, most vendor equipment has only been tested in laboratories, with few large-scale commercial deployments, and the maturity of the equipment still needs to be verified.
In addition, Zhang Chengliang pointed out that operators' insufficient understanding of how to use MSTP to provide services is also one of the reasons hindering its large-scale application. Traditional SDH equipment serves only as a transmission node in the network and cannot directly provide services. MSTP, however, is different: by using MSTP's Ethernet L2/RPR/MPLS technologies to build a metropolitan area private line service transmission network, it can provide users—especially major customers—with highly profitable private line services, which is undoubtedly very attractive to operators. However, how to establish an effective business model to gradually shift users' full confidence in the original DDN/frame relay/2M private lines to MSTP-based Ethernet private line services still requires further exploration.
Despite some problems, MSTP has already passed its most difficult period. Chen Xiaofeng, Deputy General Manager of the Optical Network Division at UTStarcom, stated that over the past few years, MSTP's functions have been continuously improved, and the technology and standards have become increasingly mature. Moreover, both traditional and emerging chip manufacturers and equipment manufacturers have invested considerable effort in the R&D and production of MSTP products, ensuring product quality while also ensuring product supply, which has driven product prices down. It is expected that the era of MSTP will arrive next year.
Several major domestic operators have also clearly stated that MSTP will be the technology of choice for their metropolitan area optical transmission network construction in the coming years, which seems to corroborate this view from another angle. Shen Wencui, Deputy General Manager of the Basic Network Department at China Unicom, told reporters that China Unicom has determined MSTP as the main direction for future metropolitan area optical network construction. Currently, China Unicom is referencing the results of the multi-vendor MSTP interoperability test conducted by China Telecom last year to formulate its own standards and prepare for the large-scale commercial deployment of MSTP. "In the future, under similar cost-performance conditions, even if the price is slightly higher, China Unicom will give priority to MSTP products," said Shen Wencui.
In addition, we can get a sense of the situation from the recent deployment of products and solutions by mainstream MSTP equipment manufacturers in operator networks: ZTE's multi-service transport platform (MSTP) ZXMP series products successfully won bids for metropolitan area network projects in Changchun, Harbin, Zhengzhou, and Tianjin; Zhejiang Telecom selected Alcatel Shanghai Bell's multi-service optical transmission solution; UTStarcom's NetRing products successfully won the 10G backbone transmission network project for Dezhou Communications Branch; Harbor Networks' OpCity series MSTP equipment was deployed on a large scale in Jiangsu Tietong's Wuxi, Yangzhou, and other city companies... Perhaps it will not be long before we see MSTP flourishing across markets everywhere.
The Future: How Should the Path to Intelligence Be Taken?
How to improve the utilization of network resources and enhance network service quality is one of the issues operators care most about. As the proportion of data services gradually increases, in order to adapt to the uncertain and unpredictable characteristics of data services, MSTP technology must further optimize the transmission mechanism for data services, gradually introduce intelligent features, and evolve toward Automatically Switched Optical Network (ASON). Pu Jian, MSTP Product Manager at ZTE's Transmission Systems Division, believes that with the development of NGN and 3G networks, service connections in the network are becoming increasingly complex, and evolution toward intelligent optical networks is the inevitable direction for MSTP.
This view has also gained widespread recognition in the industry. So how should MSTP's path to intelligence be taken? Pu Jian believes that intelligent optical network functions will first be implemented at the core backbone layer, with automatic service scheduling and automatic restoration as the primary goals. The industry holds different views on this as well. Tao Hucheng, Deputy Chief Engineer of the Optical Network Products Division at Harbor Networks, believes that considering the complexity of traffic and flow directions at the core convergence layer, as well as the importance of services and network layers, the GMPLS protocol still needs to be tested over time in terms of both security (the capability of 50ms protection switching time) and maturity (interconnection and interworking of control plane and management plane in a multi-vendor environment). Therefore, Harbor Networks adopts a cautious attitude toward the intelligence of the core convergence layer. Implementing the intelligent evolution strategy of transmission equipment starting from the network edge may be a more feasible approach, validating the technology with lower risk and thereby effectively building large-scale transmission networks.
Regardless of the path taken toward intelligence, it is certain that next-generation MSTP will integrate standards such as VCat, GFP, LCAS, and ASON, enhancing data service processing capabilities while adding intelligent features, thereby rapidly responding to real-time bandwidth demands of the service layer and providing support for operations such as bandwidth leasing, Optical Virtual Private Network (OVPN), and SLA.
We regard MSTP as the future path of the SDH network, but what about the future of MSTP itself? Will it be merely a "dazzling moment" and a "fleeting flame across the sky"? Everything remains to be tested by time.