One of the primary management challenges facing IT organizations today is the fact that in the vast majority of instances when the performance of an application is degrading, the degradation is noticed first by an end user and not by the IT organization. To further complicate the situation, when an end user informs the IT organization that the performance of an application is degrading, the organization is typically not aware of the cause of the degradation. In addition, as pointed out in Volume III of Performance-First Insights, most IT organizations work on the premise that if the performance of an application is degrading, the cause of that degradation is the network. The combination of these factors results in the tarnishing of the reputation of the IT organization and an elongation of the time it takes both to identify the cause of degraded application performance and to remedy the situation.
Unfortunately, within most IT organizations there are two fundamentally different schools of thought as to how to respond to this challenge. Individual contributors who work either in the NOC or in the network engineering and design function are the advocates for one school of thought. These individuals have a clear understanding of the complex problems they face and the level of information they need to resolve them. As such, these individuals, especially network engineers, use a wide range of management tools, each of which is narrowly focused on a particular facet of the management spectrum.
The typical IT manager, however, sees a number of issues associated with the individual contributor’s school of thought. One issue is that implementing a wide range of tools is costly, from an initial acquisition and ongoing maintenance point of view. Another issue is that having a wide range of tools leads to a further hardening of the organizational silos that exist within most IT organizations. That follows because having a wide range of tools increases the likelihood that the individual sub-groups within the IT organization will have differing, and sometimes conflicting data about the overall health of the IT infrastructure and the applications that transit that infrastructure.
In order to overcome these issues, the manager’s school of thought is to implement as few management tools as possible, with the ideal being the implementation of a single manager of managers (MOM). A MOM is a high-level network management system that gathers and correlates alarms and alerts from multiple element management systems1. In theory, this school of thought avoids the issues discussed in the preceding paragraph because it presents the users with a single view of the IT environment. Unfortunately, attempts over the last decade to implement a MOM have been less than successful in large part because there has not been a single tool in the marketplace that provides the individual contributors with all the information, management controls, and level of detail they need to do their jobs.
This edition and the next edition of Performance First Insights will provide some guidance on how to minimize the conflict and bridge the gap between the individual contributor’s and the manager’s differing schools of thought. To achieve this goal, IT organizations need to take a fundamentally new look at the MOM.
To assess the current management environment, it's instructive to review the results of a survey that was administered to 176 IT professionals in 2008. Throughout this document, those 176 IT professionals will be referred to as The Survey Respondents. Roughly half of The Survey Respondents work in a NOC. The other half of The Survey Respondents work in IT, but not in a NOC.
As recently as a few years ago, NOC personnel spent the majority of their time focused on traditional fault management tasks such as ensuring the availability of switches and routers. To understand how this allotment of time might have changed, The Survey Respondents were asked to indicate the component of IT that consumes the greatest amount of time on the part of NOC personnel. Their responses, which are shown in Table 1, indicate that over the last few years the role of the NOC has fundamentally changed. These responses also indicate that the changing role of the NOC is not well understood inside the rest of the IT organization.
In particular, The Survey Respondents who work in the NOC are in the best position to know where NOC personnel spend their time: roughly forty percent indicated that NOC personnel spend the greatest amount of their time focused on applications. This was almost twice as large as the next closest response: NOC personnel spend the majority of their time focused on the WAN. The responses of the NOC personnel are in sharp contrast to the responses of the non-NOC respondents: roughly half of the latter respondents think that NOC personnel spend the greatest amount of their time focused on the WAN and relatively few of them think that NOC personnel spend the greatest amount of their time focused on applications.
| Work in the NOC | Do Not Work in the NOC | |
| Applications | 39.1% | 17.2% |
| WAN | 23.4% | 48.3% |
| Servers | 14.1% | 22.4% |
| LAN | 10.9% | 3.4% |
| Security | 9.4% | 5.2% |
| Storage | 3.1% | 3.4% |
It is difficult for any organization to be regarded as being successful if its role is not well understood, which partially explains why so many NOCs are not currently regarded as being successful. For example, The Survey Respondents were asked a number of questions about how the senior management of their IT organizations view their NOCs. Over a quarter of The Survey Respondents indicated that their senior management does not believe the NOC is meeting their organization’s current needs. This lack of confidence on the part of senior IT managers is one more compelling reason why IT organizations need to make a fundamental change in how they approach management in general, and how they think about a MOM in particular.
The good news is that most IT organizations know that they need to make changes to their NOC. In particular, two thirds of The Survey Respondents indicated that their company will make changes to their NOC within the next year. The data in Chart 1 shows the factors that are driving change within the NOC.
The top five drivers of change listed in Chart 1 are also the primary characteristics of the new age MOM necessary for NOCs to meet both current and emerging requirements.
The IT management function is evolving away from an approach that focuses on individual technology domains (e.g., LANs, WANs, servers, data bases, firewalls, etc.) and is adopting an approach that focuses on the experience that a user has with the company’s key applications. One of the long-terms goals of this evolution is for IT organizations to create the management environment in which, before the performance of an application severely impacts end users, the IT organization is aware of the situation and has identified the root cause of the problem.
While that goal may not always be attainable in the near term, there are some short-term goals that are attainable. Those goals are:
For the reasons previously mentioned, managers are attracted to the concept of a MOM as a means to improve their management capabilities, and it’s a powerful concept. However, the reality is that traditional MOMs have not provided the broad set of performance metrics and level of detail that individual contributors need. In addition, the complexity of the typical IT environment is increasing due to the deployment of new application architectures such as Services Oriented Architecture (SOA) and new technologies such as virtualized servers and virtualized desktops. As a result, individual contributors will need new and even more detailed management data to do their jobs.
As they strive to realize the short-term goals listed above, IT organizations should not avoid the concept of a MOM, but they should work with their management vendors to insist on a change relative to how a MOM is designed and implemented. For example, the traditional concept of a MOM is predicated on tight integration of the management functionality provided by a single vendor. Since no vendor will always provide best-of-breed solutions for all aspects of management, the traditional MOM has failed to meet all the needs of individual contributors. The way to overcome this limitation is to deploy a new age MOM – one that is predicated on the tight integration of best-of-breed functionality, regardless of supplier.
The need for this type of integration is demonstrated by the data contained in Chart 1 which shows the need to improve communications both within the NOC, and between the NOC and other IT functions, such as network engineering and design. One way to improve communications is to consolidate management tools. By doing this, personnel in the NOC and in the network engineering and design function have the same view of the IT environment.
The data in Chart 1 also shows the need for IT organizations to focus on application performance while simultaneously getting better at troubleshooting. This drives the mandate for IT organizations to insist that their vendors develop a new age MOM that is based on the tight integration of best-of-breed application delivery management (ADM) functionality with best-of-breed fault management. The ADM capability of such a MOM enables IT organizations to monitor the key performance metrics of the company’s primary IT resources in order to anticipate application degradation. The fact that the ADM functionality is tightly integrated with the MOM’s fault management functionality means that the MOM not only identifies when the application is degrading, but also the root cause of that degradation.
In order to achieve the long and short-terms goals described above, IT organizations must implement a MOM that is based on the tight integration of best-of-breed application delivery management (ADM) functionality with best-of-breed fault management. However, any successful management system is a combination of tools and processes. With that in mind, the next edition of Performance-First Insights will discuss the information the MOM must provide and will outline what new workflows need to be implemented.
CA | NetQoS - Network Performance Management products and services for the world's largest networks. © 2001-2010 CA | NetQoS, Inc. All rights reserved.
Newsletter Sign-Up