| We briefly discussed the alliance between IBM and Cisco in Chapter 3. IBM and Cisco are collaborating on a set of open software technologies designed to increase the end-to-end intelligence and responsiveness of the global IT infrastructure—representing a major advancement in the development of "self-healing" computing systems and networks. IBM and Cisco are working on a set of proposed technologies and standards creating a common language to detect, log, and resolve system problems. Cisco started developing its own version of autonomic computing and called it the Adaptive Services Framework (ASF). It soon became apparent to senior Cisco management that there was a good deal of synergy and overlap with IBM's autonomic computing projects, and it made good sense to think about a joint venture. ASF is a series of proposed interfaces and formats that allows customers to interact with service providers. ASF enables automation detection, diagnosis, and rectification of a wide range of systems problems and support services. ASF is leveraged on autonomic aspects of self-configuring, self-healing, self-optimizing, and self-protecting and is being developed along the same processes. Cisco intends to integrate these technologies into its products and services. Together, the system and network enhancements will help to enable a self-healing enterprise infrastructure. Pioneered in the development labs of both companies, the new problem determination technologies are envisioned as the basis for standardized exchange of problem determination data across the IT enterprise. These new technologies lay the foundation for systems and networks to detect, analyze, correlate, and resolve IT problems and automatically diagnose the root cause of a problem in complex systems. Customer datacenters have a large set of products with diverse and often proprietary instrumentation, which makes end-to-end analysis difficult or impossible. Building and implementing networked self-healing systems into the enterprise environment is a gradual process, one IBM has already begun helping the industry to undergo, by developing a common approach and terminology to architecting autonomic computing systems. One of the factors contributing to the complexity in problem determination is the multitude of ways that different parts of a system report events, conditions, errors, and alerts. Today, even simple e-business solutions may contain as many as 25 to 40 logging mechanisms. These log files contain a variety of content formats because systems are built using disparate pieces and parts, often with products from multiple vendors. Technology developed by the IBM has been crucial in overcoming these difficulties by providing a way to automatically learn the format of log files, thereby reducing the time required to process and develop interfaces to new log data (often from days to hours). The network has always played a pivotal role in enabling companies to automate their business processes. As the network increases in strategic value, the intelligence of the network becomes the determining factor in customer success. Cisco continues to lead in the creation of integrated intelligent networks through the delivery of faster, smarter, longer-lasting systems, services, products, and technologies. During this process, businesses can expect to see value on numerous levels, including a reduced total cost of ownership, and improved availability and productivity. The two companies draw on their strengths in Internet infrastructure, e-business systems, networks, and services to deliver end-to-end Internet business solutions to enterprises and service providers. Today's Network Challenges On the networking side of IT, increased complexity has resulted in many significant challenges, especially in network software products, protocols, and services. This is caused by the rapid implementation of thousands of network products and features in combination with high transaction rates in multiple vendor environments. One basic problem is the understanding of how these products work and behave, and the results that they give. ASF can address this problem by providing a range of tools, from simple network management to advanced monitoring, and analysis of problems and behavior. ASF is the superglue that joins the front and back end systems together. Figure 14.4 describes the logical progression of the Adaptive Networking Progression. Figure 14.4. The stages of Cisco's Adaptive Networking Progression. Note the similarity to the levels of autonomic computing. 
The four levels can be described as follows: -
The first level is the active stage. Here the network has the capability of addressing and communicating with devices and protocols. -
At the reactive stage, the network can react to problems and take steps for remediation. ASF is considered to be reactive at this stage. -
At the predictive stage, the network is monitoring and making intelligent decisions according to predetermined service level agreements. -
The last stage, adaptive, builds on the previous three stages, so the network can be optimized, automated, and self-managed. The Adaptive Network Framework—Principles The Adaptive Network Framework (ASF) consists of a number of principles, which have the characteristics of elements, rules, ontologies, and methodologies. The basic core functionality includes: -
Mechanisms for diagramming models of the network, devices, and servers in an autonomic format. -
Common Event Format, for events from a device to an adaptive or autonomic manager or network management system as well as those shared between agents and the network management system. -
A common format to analyze events and discard unwanted events. -
A common format that can describe symptoms and problems based on actual network events. -
Standardized message formats for exchange with operational autonomic or adaptive elements. -
Widespread adoption of industry-based open standards. Conxlusions The development of autonomic management products in the network arena is needed. The timing of this joint venture is significant as this work is resulting in both products and services for both companies can only benefit the IT industry. This venture brings together two large vendors who have substantial resources to make it a success. We need to see more joint cooperation between Zother major vendors in the self-managing market. |
