|
Odessa
It's a crucial requirement for application middleware platforms to optimise the utilisation and management of IT deployments, and enhance the quality of service (QoS) of business-critical applications automatically. Odessa project develops work load management technology for middleware to supporting different workloads with different QoS constraints. The Odessa workload management consists of the admission, queuing and dispatching of incoming requests according to operational policy. In optimising admission rates, queue lengths and dispatch rates, several factors are considered, including service policy, CPU and memory consumptions, session structures, weights and load balancing. Besides supporting traditional transaction-oriented HTTP workload, this project developed methods, algorithms and software components specifically at the network edge to deliver high quality SIP services. Moreover, support for new types of workloads and runtime environment are explored, i.e., complicated mixing of different kinds of workloads, such as the workloads of Web 2.0 applications or converged applications; complex runtime environment caused by server consolidation through virtualisation.
SIP off-loading engine (SOE)
SOE(SIP off-loading engine) accelerates SIP protocol processing by offloading time consuming components ( transport, parser and transaction ) from SIP server to front-end SIP accelerators. The implementation of SOE has demonstrated its efficiency by improving SIP server performance for about 60%. SOE helps IBM establish marketplace leadership in SIP application server with leading performance.
Datapower SIP Appliance
The goal of this project is to architect and develop an appliance for converged services in next generation networks. The appliance will address the emerging workloads in a converged network environment, bridge the enterprise SOA with SIP applications for IT and telecom service integration. We leverage existing DataPower XML appliance platform and extend it with SIP mediation, security, workload offloading and acceleration. Appliance performance optimisation on x86 platform and PRISM-like platform are also important content of this project.
Real time collaboration
Integrated IP communication has been a productivity improver with cost reduction for enterprises and organisations. Working with VoIP infrastructure, it can provide pervasive access for employees within a single unified client. Integrated IP communication solution built by IBM CRL provides two key services, one is Click-to-Call, and the other is Federated Presence. With Click-to-Call, the user can setup audio/video calls simply by a mouse click. Federated Presence can show the information for multiple terminals owned by the peer in a unified client, e.g. Sametime. Working together, they can help the user contact the right person at the right time through the right approach. As the next step, we will focus more on enterprise domain to streamline human interactions (customers, partners, employees) through seamless integration of communications capabilities with enterprise business processes.
Web 2.0 content syndication
Web 2.0 Content Syndication project aims to create reusable functions for disseminating change information of telecom / enterprise data across network boundaries. It supports various ways to publish data updates to web clients, like web feeds or JSON encoded Bayeux events. It also provides general API to data owner systems for creating change sources of data and post changes and offers management functions to control data update access. With the content syndication solution, telecom operators/enterprises can have a better communication pattern with subscribers, offer differentiated mash-up services, and enjoy easier collaboration with third party service providers.
Service Control Layer (SCL)
The Service Control Layer (SCL) project aims to provide a flexible middleware based approach to enable telecom service providers to apply flexible and transparent control over converged IN/IP, telecom/Web, telecom/enterprise service, which meet rapidly changing business and market requirements and speed time to market of new converged services. Furthermore, the monitoring and control of SCL is online and real time.
OSS New Service Impact Assessment (NSIA)
Rapid growth of network and services lead to a big challenge of performing Service Maintenance and Quality Management in diverse deployment environment. Currently, although there are already some commercial products which address management issues through physical layer to application layer of network, but it is still expensive to solve the cross-layer maintenance problem and to indicate the root cause of quality degradation in the realistic network. Service Provider needs a full-dimensional understanding of not only network/service they managed but also the quality they provided.
OSS NSIA (New Service Impact Assessment) project focuses on cross-layer correlation and characterisation of management information through System Base-lining and Situation Awareness technology and tries to address the problem of how to show the operators the insight of network through semantics situation rather than tons of obscure raw data.
Template Service Creation Environment (SCE)
With the boom of telecom value-added services, service providers are facing the challenge of how to reduce the development cost and time to market of new services. It's expected the service creation time can be reduced to 1 or 2 orders of magnitude, from months or weeks to days, even hours so as to catch many market opportunities. After studying on real running services, the service templates are used to model the service logic, process flow, telecom capability usage and expose the configurable points for LOB to create services. The differentiations between services with similar model can be easily identified by lines of business. The technical impacts of this project are defined general template model and built methodology to create reusable template for new services. It will utilise these features to provide a LOB oriented SCE to cover creation of about 60% current telecom services in days.
Service differentiation
Telecom operators are transforming their silo service infrastructure to the integrated framework in order to shorten time to market of new services and decrease the operation cost. SDP&OSS/BSS (Service Delivery Platform & Operation Supporting System/Business Supporting System) Integration layer is built to handle interaction between all silo service platforms and OSS/BSS systems to support Fulfillment, Assurance, and Billing processes for telecom value-added services (VASs). As value-added services, business processes and involved customers are of different business value, so service differentiation is an important feature in SDP&OSS/BSS integration layer to maximise total real business value of VASs.
This project aims to develop a scheduling mechanism to enable value based service differentiation to process services during business process execution. The objective is to maximise customer value which has dynamic relationships with customer satisfaction and service quality.
Real-time Service Campaign
One of the challenges for Telecom operators to be competitive comparing with Internet service providers in offering innovative applications is that their campaign management mechanism is designed for traditional voice services and less efficient in supporting vast number of value added services. A successful service campaign of value-added services must:
- Understand customer needs at a small user community
- React quickly to competitor campaign with compelling offers
- Identify and execute service campaign and offer most suitable service package in ahead of the competitors
CRL team is developing the key technology to support timely and accurate service campaign, which includes:
- Telecom Service Campaign Creation and Mgmt Environment which is a Web based environment to help market staff to design and manage service campaign based on templates
- Telecom Value-added Service Analysis Model which extend IBM telecom industry model to provide a set of model specifically supporting VAS marketing & sales
- Recommender System which is a Hadoop based distributed recommendation engine which can recommend services, contents to mobile user
SpamWatcher
It is a solution which leverages streaming processing (temporal feature) and social network analysis (spatial feature) to filter Spam SMS senders. The streaming and social network information is generated from online operation system simultaneously. The detection is completely online and real time. The technology can also be applied to other areas of fraud detection in telecom industry, like spam call.
MobiSphere
The emergence and deployment of broadband wireless networks, new types of mobile phones, mobile web technologies and market demands are putting mobile on a collision course with the internet and will drive significant industry transformation of the telecom and internet service providers and enable significant new application growth. New technologies need to be invented to support new mobile web applications, which include platforms for mobile application development, execution and management, mobile device management and mobile security. MobiSphere is a cross lab research program which aims to develop the enabling technologies. Researchers from IBM China Research Lab, Haifa Research Lab, India research Lab, Tokyo Research Lab and Watson Research Center are working on developing a platform to support the enterprise use of the mobile platform. CRL contributes BCIM (Business Communications Interaction Manager) to MobiSphere, which enables the seamless integration of mobile communications capabilities with enterprise business processes. BCIM bridges the enterprise business processes and the underlying communication infrastructure. It translates the high level business interaction requirements into micro-flows of communication capabilities, interprets the communications events initiated from the network side, and maps the events with relevant processes to drive the process to move forward.
Mobile Cloud
Cloud Computing is a new distributed computing paradigm where services are delivered by centralised data centres over the web to large scale users in a cost efficient way. Mobile Cloud is looking at: a) enabling efficient access to Cloud services via mobile devices, b) leveraging Cloud techniques to make mobile services extremely scalable and low cost. We are researching on the next generation mobile network infrastructure which delivers voice, video, collaboration and other type of services in a cloud way.
|