An action is an automatic response to an event, based on a policy. There are various types of actions including sending alerts, taking diagnostic snapshots, remediation through scripts, cloud auto-scaling, or custom.
An alert notifies a recipient list of a problem or event; by email, SMS or customized to interface with external notification systems.
An alert digest compiles alerts and sends the compilation sent by email or SMS to a recipient list at a configured time interval.
Anomaly detection refers to the identification of metrics whose values are out of the normal range, where normal range is based on dynamic baselines that AppDynamics has created based on the previous performance of these metrics.
An Ajax request is a request for data sent from a page using the XHR API. This API is used to send HTTP or HTTPS requests to a web server and to load the server response data back into the requesting page. The Ajax request is tracked as a child of the requesting page using EUM.
Application Performance Management
Application performance management monitors and manages the performance and availability of software applications in a production environment, focusing on relating IT metrics to business values.
According to Gartner research, application performance management includes: end user experience monitoring, application runtime architecture discovery and modeling, business transaction management, application component deep-dive monitoring, and application data analytics.
An app server or application server provides software applications with services such as security, data services, transaction support, load balancing, and management of large distributed systems. Examples are a Java Virtual machine (JVM) or a Common Language Runtime (CLR).
An application agent or app agent monitors an application server or CLR. App agents report data to the Controller. See Node.
See Business Application.
The application dashboard graphically represents high-level structural and status information for a single business application. The application dashboard includes a flow map, grid view, summary of key performance indicators.
Average Response Time
Average interval between the time the user request is received by the application server and the time that the response is returned to the application server. Does not include the network time for the request to reach the server or the time for the response bytes to reach the caller. Different from End User Response Time.
A backend is a software component that is not instrumented directly, but its traffic flow can be monitored. Databases and remote services such as messaging queues are collectively known as backends.
A background task or a batch job is a scheduled program that runs without user intervention.
A baseline provides a known point of reference against which performance is measured. Dynamic baselines are based on observed performance over time. Static baselines are based on specific values.
A baseline deviation is the standard deviation from a baseline at a point in time. It is represented as an integer value. Baseline deviation can be used to configure health rule conditions based on the number of deviations. For example, you can configure a warning condition as 2 standard deviations from the baseline and a critical condition as 4 standard deviations from the baseline.
A baseline pattern defines the base time period of data used to create baselines. It can be a fixed time range or rolling time range, in which the most recent x number of days is always used.
See Bytecode Injection.
A browser snapshot presents a set of diagnostic data for an individual base page, iFrame or Ajax request. The data reports the end-user's experience starting with the Web browser. Browser snapshots are taken at periodic intervals and when certain performance thresholds are reached.
An AppDynamics business application models all components or modules in an application environment that provide a complete set of functionality. AppDynamics correlates the components' activities to provide performance data.
A business metric captures data from a method's parameters or return values to report on the performance of the business, such as revenue per transaction or number of orders. Business metrics are implemented using Information Points.
A business transaction represents an collection of user requests that accomplish a logical user activity, such as log in or check out, across a distributed business application. A single request is a business transaction instance.
Bytecode injection modifies a compiled class at runtime by injecting code into it immediately before it is loaded and run.
A call graph represents the calling relationships among subroutines in an application. It makes up a part of a transaction snapshot that is used to identify root cause of a performance problem.
A compute cloud delivers computing and storage capacity as a service. Examples are Amazon EC2, OpenStack, etc.
Detection is the process by which AppDynamics identifies a business transaction or backend in a managed application. Detection is also referred to as discovery.
A diagnostic session is a session in which transaction snapshots are captured, with full call graphs. A diagnostic session can be started manually through the user interface or configured to start automatically when thresholds for slow, stalled, or error transactions are reached.
A distributed application runs on multiple computers in a network. Some of the computers can be virtual machines.
End User Monitoring
End User Experience Monitoring (EUM) provides performance information from the point of view of the client, whether that client is a web browser or a mobile native application. This is different from other types of AppDynamics monitoring, which typically begin at the application server. EUM collects a different set of metrics than the server-side app agents.
End User Response Time
Average interval between the time that an end-user initiates a request and the completion of the page load of the response in the user's browser. In the context of an Ajax request, the interval ends when the response has been completely processed. Not to be confused with Average Response Time.
An entry point is a method or operation in your application code that begins or extends a business transaction. AppDynamics automatically detects business transaction entry points for common frameworks, and you can configure entry points to customize how AppDynamics detects business transactions.
An error in AppDynamics indicates an unhandled exception in the context of a business transaction, a logged error of the appropriate severity, or any exception called during an exit call, which prevents the transaction from working properly.
An error transaction is an error that occurred during transaction execution. An error transaction can be caused by a logged error or a thrown exception.
An exception is a code-based anomalous or exceptional event, usually requiring special processing. It can occur in the context of a business transaction and outside of a business transaction
See End User Monitoring.
An event represents an action or occurrence detected by the system that can be handled by the system. There are different event types.
An exit point is a call from an app server to a backend database, remote service or to another app server. AppDynamics automatically detects many exit points and you can configure custom exit points.
A flow map graphically represents the tiers, nodes, and backends and the process flows between them in a business application.
Health in AppDynamics refers to the extent to which the application being monitored operates within the acceptable performance limits defined by health rules. Health is indicated by a green/yellow/red color scheme.
Health rules allow you to select specific metrics as key to the overall health of an application and to define ranges for acceptable performance of those metrics. You can customize default health rules and you can create new ones.
Health Rule Violation
A health rule violation exists if the conditions of a health rule are true.
High Availability (HA) Cluster
A cluster of computers that hosts duplicate server applications with the purpose of reducing down time. The HA cluster, also called a failover cluster, is enabled by redundant systems that guarantee continued delivery of service during system failure.
A histogram is a graphical representation of the distribution of data, shown as adjacent rectangles, erected over discrete intervals (bins), with an area proportional to the frequency of the observations in the interval.
The Web page you see when you first log into the AppDynamics Controller, before you have selected an application. See AppDynamics Home Page.
An iframe is an "inline frame", an HTML document that is embedded in another HTML document. It is tracked as a child page using EUM.
An information point instruments a method in application code outside the context of any business transaction. It is used for monitoring the performance of the method itself or for capturing data from the method's input parameters or return value.
Instrumentation is the mechanism by which performance data is gathered. Agents instrument Java or .NET applications. See Bytecode Injection.
Key Performance Indicator
Key performance indicators are main metrics that an organization uses to measure its success. In AppDynamics, the key performance indicators are assumed to be load (number of calls and calls per minute), average response time, and errors (number of errors and errors per minute.)
A machine consists of hardware and an operating system. It hosts application services and it can be virtual.
A machine agent instruments a machine to report data about hardware and the network to the Controller. AppDynamics provides both a Standalone Machine Agent and an embedded machine agent in the App Agent for .NET. The Standalone Machine Agent functionality can be extended to add additional metrics. The embedded .NET Machine Agent can have additional performance counter metrics.
A managed application is an application with servers that are instrumented by AppDynamics.
A match condition frames a test consisting of a match criterion (such as a method name, servlet name, URI, parameter, hostname, etc.), a comparison operator typically selected from a drop-down list, and a value. Used in many types of AppDynamics configuration to specify entities to be included in or excluded from monitoring.
On-Premise refers to an AppDynamics Pro installation where the controller is installed on machines at your site. Alternatively, AppDynamics offers AppDynamics Pro as an SaaS.
A pageview is an instance of a web page being loaded into a Web browser.
A policy consists of a trigger based on events and an action respond to the event. A policy provides a mechanism for automating monitoring, alerting, and problem remediation.
A remote service provides a service to a distributed application outside of the JVM or CLR. Examples are a Java Message Service or Web Service. See Backend.
The AppDynamics REST API is implemented using Representational State Transfer (REST) Services. You use the REST API to retrieve information from AppDynamics programmatically.
A request is a single instance of a business transaction; for example, 500 requests per minute for the "Checkout" business transaction.
SaaS is an acronym for Software as a Service. AppDynamics provides AppDynamics Pro as an SaaS where the controller is hosted on AppDynamics in-house machines and monitors your applications, communicating over the internet with app server, Java, .NET, infrastructure, and database agents installed in your environments. You can, alternatively, install the AppDynamics Pro controller on your own equipment, an installation type referred to as On-Premise.
A visual summary of the performance of a business transaction within a specified time range, covering percentage of instances that are normal slow, very slow, stalled or errors.
Service endpoints provide a subset of the metrics for a tier, focused on a specific application service. You configure entry points for the app service and AppDynamics provides metrics and associated snapshots specific to the service.
A task codifies a unit of work as a set of instructions and is a component of a step in a workflow.
A tier represents a key service in an application environment, such as a website or processing application. A tier is composed of one or more nodes or backends. An "originating tier" is the tier that receives the first request of a business transaction. A "downstream tier" is a tier that is called from another tier. See Node and Backend.
Tag, trace, and learn
Tag, trace, and learn is a methodology used for tracing code execution and discovering the business transaction context.
A threshold is a configurable boundary of acceptable or normal business transaction or background task performance.
Tracing is following the execution of software code and recording information about the execution, usually for debugging or performance monitoring purposes.
Transaction correlation is the internal mechanism that allows AppDynamics to do transaction tracing in modern web applications, tying together distributed components into a single entity, the business transaction, for monitoring purposes.
A transaction snapshot is a set of diagnostic data for a business transaction instance across all app servers through which the business transaction has passed, at a specific point in time. Transaction snapshots help you troubleshoot the root causes of performance problems.
The process of using a dynamic value to customize how Business Transactions are identified.
A workflow builds a sequence of steps in which each step consists of one or more tasks that are executed on a machine instrumented by a Standalone Machine Agent.