Define, Deliver, Assure

Service Lifecycle Vision Invades Service Provider OSS

Service providers, seeking a return on investments in trenches filled with fiber optics and data centers with servers, are racing to launch managed security, storage, business application and other value-added, premium-priced services.

They find themselves competing in an environment where the most demanding top 20 percent of customers may account for 80 percent of revenue, and the provider that can define, deliver and refine new services faster than the next guy is likely to win.

Consequently, some providers have begun to employ a new generation of operational support systems (OSSs) that apply an automated lifecycle approach to service definition, activation and assurance.

The goal, says Deepak Swamy, chief strategy officer for OSS vendor Trendium Inc., is to turn the service provider's network operations center (NOC)--which is concerned only with one or another set of devices--into a "service operations center, or SOC." The SOC calls on centrally stored service definitions, user profiles and policy applications to manage every layer in the service topology.

The SOC concept allows:

• Correlation among repositories of delivery resources, service profiles and customer profiles;

• Automated assembly of service components;

• Rapidly launchable, pre-architected combinations of those components;

• Service selection portals that allow customers to further personalize those combinations themselves;

• And, service policy administration and controls to enforce performance down to the individual session, service and user.

This goal is echoed by network and data center OSS system vendors, including Atreus Systems Corp., CoManage Corp., Dorado Software Inc., Ellacoya Networks Inc., Emperative Inc., Redback Networks Inc., Sphera Corp., Thor Technologies Inc., Trendium, Unisphere Networks Inc., Viewgate Networks Inc., XACCT Technologies Inc. and Xevo Corp.

Xevo's XevoLogic engine combines three modeling processes: (1) a service model, to build application, storage, quality of service level, help desk level and other components of a service; (2) a hosting model, to define server capacity and other supporting infrastructure requirements; and (3) a subscriber model, to define customer organizations, departments, users and their access profiles.

"The subscriber profile applied to the hosted service model is what informs the provisioning," says marketing vice president Greg Kee. "That's what enables point and click service packaging."

Providing OSS to customers such as IP Communications Dorado Software addresses the issue from the networking point of view. "We specialize in dynamic, changing factors including users, network resources, service levels, time of day, type of content, security breaches," says founder and CEO Tim Sebring. "From creation through provisioning through service level management, customers want to change these things dynamically."

To close the lifecycle loop, says CoManage business development vice president Razi Imam, requires solving the "fragmentation" between customer- facing OSS silos, such as order management, customer care and billing, and network-facing OSS silos including asset discovery and inventory, provisioning and fault management - as many as 13 management points for network operations center personnel.

In contrast, he says, providers including AT&T Corp.and Cox Communications Inc. are using CoManage's Integrated Services Manager as a single OSS application to communicate with all those systems for rapid creation, launch and assurance of tailored services.

For Trendium, whose ServicePATH OSS customers include managed hoster NavLink Inc., the vision is that building blocks-virtual private networks, applications, storage-can be assembled into layered services, a goal hindered by the isolation of frame relay, Internet, data center and other OSSs from one another.

"To evolve from traditional, one-to-one outsourcing to replicable services," Swamy says, "you have to encapsulate best practices and processes into a service creation, provisioning and assurance system, and then use that system to manage and enforce those processes."

In this scenario, the enforcement becomes both end and beginning of the cycle, enabling a service provider to "know the quantifiable throughput, availability and other business objective capabilities at each layer," he says. "Those capabilities must be tied closely to defining the service, which includes rules for what to do when a service level fails." Given such a failure, he adds, "the correlation to the living, electronic contract can be automated."

Paradise Lost

Service lifecycle integration's ultimate power will lie with its ability to enable service providers to exploit what CIMI Corp. estimates will soon be a $300-billion content and application service opportunity on top of another $300 billion in transport and access services.

To avoid missing that boat, says Laurent Delifer, president and COO of NavLink--a provider of managed server, VPN, firewall, intrusion detection and other services to customers of partners AT&T and British Telecommunications plc -- installed Trendium's ServicePATH earlier this summer. Now, he says, introducing new services "is much easier, and we can still customize."

"You have to be able to create new services in response to actual demand, and you have to be able to deliver them immediately," says Virginia Brooks, director of product marketing for Unisphere Networks Service Selection Center system, now in the labs of some 35 service providers, including Sprint Communications Co. and Deutsche Telekom. "Otherwise, the service provider is leaving a lot of lost opportunity on the table."

Share this article: Email, Slashdot, Digg, Del.icio.us, Yahoo!MyWeb, Windows Live Favorites, Furl
Add this article feed to: RSS, My Yahoo, Newsgator, Bloglines