End-to-end APM Overview - What you need to know

W H I T E PA P E R
BEST PRACTICES: APPLICATION PERFORMANCE MANAGEMENT
OVERVIEWW H I T E PA P E R : A P P L I C AT I O N P E R F O R M A N C E M A N A G E M E N T
Best practices for Application Performance Management
End-to-end Application Performance Management The APM Best Practice series will have three more papers following this overview, focusing on the key areas that impact overall service delivery:(APM) refers to the IT service discipline that . Datacenter best practices: a look at how APM delivers true end-to-encompasses the identi? cation, prioritization and end application service for heterogeneous environments, including resolution of performance and availability problems distributed and mainframe.affecting business applications. APM is increasingly . Network management best practices: a look at application performance management in a network-dependent application important as end-users rely on ever-more complex delivery model. applications to enable critical business transactions. . Java and .NET best practices: a look at application performance Poor application performance reduces productivity, management in a J2EE and .NET environment.impacts customer satisfaction and damages IT's As preparation for these upcoming papers, we'll ? rst provide an overview of APM design, implementation and operation fundamentals, discussing reputation, resulting in increased costs, lost revenue end-to-end APM as a process.and IT inef? ciency-often to a greater degree than availability problems. APM DESIGNTraditional monitoring solutions often fail to identify and solve the root APM solutions often begin as grass-roots, infrastructure-monitoring causes of application performance problems. In fact, recent advances exercises, implemented by individual silos within the IT organization and in end-user experience monitoring, dependency mapping and correlation lacking cohesive goals. For example, the network team may implement have enabled IT operations managers to monitor and troubleshoot service an open-source network tool to gain basic network visibility, while the level breaches more effectively. These technologies help increase the web server team may implement a server-monitoring tool from a popular visibility across the network, server (distributed and mainframes) and server vendor. However, it is much more practical to design an APM other application tiers, adding technical insight into cause and effect, solution from the top down, using a service-oriented approach that breaks as well as business insight into response prioritization. In effect, the down the walls between IT organizational silos. Using this approach, you emphasis has shifted from infrastructure metrics to business metrics, envision the outcome ? rst, and then apply it to your choices of solution even though infrastructure metrics continue to provide the underlying components. fault and capacity data. Where do you begin? In the ITIL world, operational level targets (OLTs), In this series of best practice papers, we explore APM with an eye ultimately supporting service level agreements (SLAs), represent a towards problem and incident management, two of the core Information good starting point; these will already have addressed the desired Technology Infrastructure Library (ITIL) processes facilitated by APM. business outcome and cost constraints, and should facilitate a high-level Incident Management addresses IT issues as they occur, offering a design. Not an ITIL shop? You can still adopt the parts of the best structured response to a reduction of service quality. The goal of Incident practices that ? t your needs. Begin with a discussion with the business Management is to restore service with as little business impact as group to understand business goals, identify the APM budget, apply possible. Problem Management emphasizes identifying and eliminating a an understanding of the application delivery infrastructure and its problem's root cause. It adds a notion of service-quality improvement via changes to the service as well as to the APM solution. 2 W H I T E PA P E R : A P P L I C AT I O N P E R F O R M A N C E M A N A G E M E N T
performance sensitivities, and draft a solution. You might consider this an Important in the design process is to keep focused on end-user exercise in examining what could possibly go wrong, expanding the scope transaction response ti... [download for more]