Service provider takeaway: Service providers can choose among an array of desktop performance monitoring tools to help customers get a better handle how end-user machines figure into the overall performance equation.
The IT services management (ITSM) movement pushes the practice of different parts of an organization establishing service-level agreements (SLAs) with one another, just as they would with external service providers. At the same time that ITSM thinking -- as exemplified by widespread use of the IT Infrastructure Library (ITIL) -- has spread widely through organizations of all sizes and industries, IT's service delivery context has shifted to a more dispersed group of end users.
Nemertes Research's "Building a Successful Virtual Workplace" report found that approximately 90% of the workforce doesn't work at headquarters, nor at the main data center site. These workers are in branch offices of one sort or another, with the number of branch offices growing by about 9% annually. However, new branches rarely have their own data centers, and old branches are steadily stripped of them as IT services are centralized. So IT must deliver more crucial services to the desktops of a steadily growing number of sites and users, which increases the importance of SLAs and focuses attention on desktop performance.
Systems integrators and value-added resellers (VARs) can help their customers glean meaningful data about the performance of client computers by providing the expertise IT needs to get started deploying and configuring desktop performance management tools, or by themselves managing desktop performance along with desktop hardware and software.
Desktop performance data gap
It's impossible to promise realistically, or deliver on promises reliably, without instrumentation that can monitor performance. Enterprise application performance depends on data center systems, network infrastructure and deskto
To continue reading for free, register below or login
To read more you must become a member of SearchSystemsChannel.com
ps, but most enterprises cover only the first two, servers and networks, and neglect the third. Nemertes' 2006 study "Delivering the Enterprise" found that only one in five companies does direct, regular assessments of application performance at the desktop.
Without looking directly at the desktop as a part of the performance equation, when a performance problem is rooted there, they can't identify it with the same ease and reliability as, say, a congested network node or overburdened server. A tech might have to take remote control, or worse, visit in person to figure out what is going on. The most popular remedy is often, still, "Reboot and call me in the morning." But as desktop performance becomes a more crucial part of the delivery of central services, this disruptive solution and the attitude and posture behind it become increasingly hard to justify.
The increasing demand for a growing array of centralized services from consolidated data centers over converged networks to a virtualizing workforce has steadily ratcheted up the dependence on IT services. But IT budgets and staffs have, of course, not grown at the same pace. IT is continually striving to do more with less and has adopted a number of strategies to cope. Some are technical, such as automation and virtualization, and some are organizational, such as outsourcing and migrating to applications based on Software as a Service (SaaS). Often, the two strategies go hand in hand: Organizations look to professional services companies to implement a new technology and to perform skills transfusions so that their own staff can manage the new tools.
As they are driven into unknown territory in order to monitor performance on the new enterprise desktop, it is natural that many organizations will be looking to their systems integrators and service providers to supply the expertise they lack. In fact, being able to provide accurate performance data on desktops could become a crucial differentiator for companies providing desktop services to enterprises. Whether it is initial deployment or ongoing management, a company with a proven toolbox for performance monitoring would have a compelling sales pitch. Being able to ensure the systems they supply can tell IT whether desktop performance is within promised limits -- or exactly how it needs to be improved to support a new initiative -- is a very valuable service.
Desktop monitoring tools and services
There are several varieties of desktop performance monitoring available, aimed at different aspects of the new enterprise desktop. Some, like Aternity Inc.'s Frontline Intelligence Performance Platform, look deep into the operating system to get answers about what within the PC operating environment is consuming resources. Others, such as Centrisoft Aware, are focused on network applications specifically. Most are agent-based, and many can provide rich data for analyzing transaction times for a broad palette of applications, both fat client and Web-based. Some might work well in conjunction with centralized appliances for measuring network transaction data within the data center, such as Symphoniq's TrueView system.
Systems integrators and service providers can provide real value to IT by handling the details of inserting agents seamlessly into the desktop environment (and possibly appliances into the network) with minimal impact on performance and maximum ability to integrate data into existing management tools and consoles, as well as expert advice on which data to pay the most attention to. Many management tools are not able to deal gracefully with the massive volumes of data that a desktop performance monitoring solution might produce, so there might be issues with folding the data into a manager of managers or other high-level monitoring solution. It might be necessary to introduce an intermediate consolidation layer, for example. Integrators and service providers can again supply the necessary expertise before, during and after rollout to ensure that the new data does not overwhelm core management tools.
No matter the context, companies providing desktop implementation and management services should build performance monitoring and management into those systems so that they can provide their customers with the richest possible information about how and how well they are performing. With sufficient granularity and reliability in that information, IT organizations will be able to incorporate accurate data on desktop performance into their business cases for new systems and services.
About the author
John Burke, principal research analyst, worked in academic IT for 18 years before joining Nemertes Research. He has worked as an end-user support specialist, programmer, systems administrator, database programmer and administrator, network administrator, network architect and systems architect. As an analyst, John draws on his experiences as a practitioner and director of IT to better understand the needs of IT executives and the challenges facing vendors trying to sell to them.
