In Datadog RUM, an action is a type of event. Action events track user interactions during a user journey.
Synonyms: admin status
The administrative status of a port (up/down) refers to whether the port is disabled.
Synonyms: Datadog Agent
The Datadog Agent is an open source software that runs on a host. It collects metrics and events from the host and sends them to Datadog. It can run on your local hosts (Windows, macOS), containerized environments (Docker, Kubernetes), and in on-premises data centers. For more information, see the documentation.
Datadog monitors generate alerts, indicating that a set condition has been reached.
ECS is a container orchestration service.
EKS is a managed Kubernetes service.
An ARN is a string that uniquely identifies an AWS resource. See the AWS documentation for more information.
Log analytics is the process of querying, grouping, and visualizing logs for investigation and exploration.
In Kubernetes, annotations are key/value maps that can be used to attach metadata to Kubernetes objects.
An API key is a token used to authenticate a user or an application. The Datadog Agent requires an API key to submit metrics and events to Datadog.
Related terms: Check
Synonyms: Tracing, Distributed Tracing
Application Performance Monitoring (APM) monitors requests, errors, and latency in your application. Add distributed traces throughout your application to correlate to browser sessions, logs, profiles, synthetic checks, network, processes, and infrastructure metrics across your hosts, containers, proxies, and server less functions.
For more information about APM, see the APM documentation.
An archive is a long-term cloud storage solution to store logs, whether they are indexed or not, for longer periods.
An attribute is a piece of information about a log.
In Datadog, Autodiscovery is a feature that automatically identifies the services running on containers. This enables you to define configuration templates for Agent checks and specify which containers each check should apply.
Synonyms: Fargate
AWS Fargate is a serverless compute engine.
AKS is a managed Kubernetes service.
In Datadog, cardinality is the number of tag values associated with a tag key for a metric.
The CIS is an organization that maintains CIS Controls and CIS Benchmarks, which are recommendations and guidelines for best security practices.
Synonyms: APM
Checks are small Python programs run periodically by the Agent. A Check performs an action and then gathers the result, which the Agent then stores and reports to the Datadog platform. These programs are freeform and are generally used to collect metrics from custom environments or applications. Note that the word “check” - when not capitalized - refers to the generic act of taking a measurement.
A child org belongs to a parent org and maintains its own data separate from the parent org and other child orgs.
The Cluster Agent is a version of the Datadog Agent that provides a streamlined, centralized approach to collecting cluster-level monitoring data.
In computing, a cold start refers to when a system or component was recently created or restarted. In serverless computing, a cold start refers specifically to the problems (such as increased latency) that may arise when a function is invoked for the first time or after an idle period.
The collector is the Agent process that runs checks on the machine and collects metrics. For more information, see the documentation.
A ConfigMap is an API object that stores data in key-value pairs. ConfigMaps can be supplied to Pods as environment variables, command-line arguments, or configuration files in a volume.
The Container Agent is the version of the Datadog Agent that runs on a containerized environment.
A container runtime is the part of a container engine that mounts the container and stops/starts containerization.
The CRI interface allows a kubelet to use different container runtimes.
Containerd is a container runtime.
Core Web Vitals are a set of metrics that Google has identified as meaningful signals for UX testing. These metrics concern how long it takes for a page to load, how long it takes for a user to interact, how stable a page is as it loads, etc.
“Count” is a metric type that adds up all the submitted values in a time interval. For more information, see the documentation.
A crawler delay is a delay in metrics for Datadog cloud integrations due to constraints with the cloud provider API. For more information, see the documentation.
Synonyms: XSRF
CSRF is a type of exploit where an attacker uses a web client, such as a browser, to access or manipulate information.
In Kubernetes, a DaemonSet is a controller that manages groups of Pods. You can describe a DaemonSet in a YAML file.
The datadog.yaml is the main Agent configuration file for enabling and disabling different features.
For more information, see the documentation.
An evaluation delay tells the monitor to wait a specified number of seconds before it begins evaluation. For more information, see the Advanced alert conditions.
Distributed tracing is a method of tracking application requests as they flow from frontend devices to backend services and databases. Developers can use distributed tracing to troubleshoot requests that exhibit high latency or errors.
A distribution is a metric type that aggregates values sent from multiple hosts during a flush interval. For more information, see the documentation.
Docker is a framework for managing containers.
DogStatsD refers to two related things: a protocol based on StatsD, and an application for reporting metrics which implements that protocol. The DogStatsD protocol is an extension of the StatsD protocol, with some modifications that are specific to the Datadog platform. The DogStatsD application is a service that is bundled with the Agent, and is used as a lightweight mechanism for reporting metrics.
See the DogStatsD documentation for more information.
Downtimes are scheduled periods during which monitors’ alerts and notifications are silenced. For more information, see the documentation.
DAST is a security testing methodology that analyzes a running application without looking at its source code.
eBPF is a Linux kernel technology that allows users to run bytecode without changing the kernel or adding kernel modules.
Synonyms: enhanced Lambda metric
Datadog generates a set of enhanced Lambda metrics from your Lambda runtime. These are in addition to the default Lambda metrics provided by the AWS Lambda integration. Enhanced Lambda metrics are prepended with aws.lambda.enhanced.*.
In Datadog RUM, an error is a type of event. An error event is generated when the browser emits a frontend error.
The evaluation window is the look back timeframe of the data that the monitor aggregates and uses to compare against the defined thresholds. For more information, see the documentation.
An exclusion filter determines which logs should not be indexed. These logs still show in Live Tail.
Events Explorer is a page in Datadog where users can view and aggregate events. Events Explorer displays the most recent events generated by the user’s infrastructure and services, such as code deployments, service health, configuration changes, or monitoring alerts. For more information, see the documentation.
A facet is a user-defined tag or attribute of indexed logs. It can be quantitative or qualitative and is used in Log Explorer to search logs, define log patterns, and perform log analytics.
A faceted search uses filters to narrow down search results.
A flame graph is a visualization of a trace, where bars represent spans and show the the span’s execution time as well as what called it and what calls it made. Flame graphs are also used to represent profiles.
The flare command is a quick way to send troubleshooting information to the Datadog support team. flare gathers all of the Agent’s configuration files and logs into an archive file, removes sensitive information like passwords, and then sends the archive file to Datadog support.
In computer networks, a flow is the path taken when one endpoint communicates with another. Datadog’s network map provides a visualization for network data flow. For more information, see the documentation.
Forecasts use algorithms to predict the future behavior and values of a metric.
The forwarder is the Agent process that sends metrics over HTTPS to Dataodog. For more information, see the documentation.
In serverless computing, a function is a programmatic function hosted on managed infrastructure.
Funnel analysis analyzes a user’s journey towards a defined outcome, such as signup or purchase. Funnel analysis looks at the sequence of events along this journey.
Gauge is a metric type that takes the last value reported during the interval. For more information, see the documentation.
In Datadog Synthetic Monitoring, a global variable is a variable that is accessible from all of a user’s Synthetic tests. For more information, see the documentation.
GKE is a managed Kubernetes service.
Granularity is the frequency at which data is collected or displayed on graphs. For more information, see the documentation.
Grok is a method for parsing and extracting attributes from semi-structured log messages.
Helm is a tool for managing pre-configured Kubernetes resources.
A histogram reports five different values that summarize the submitted values: the average, count, median, 95th percentile, and max. For more information, see the documentation.
In Kubernetes, an HPA automatically deploys more Pods to meet demand.
A host is a computer or a virtual machine.
Indexed logs are logs that have been collected, processed, and retained for analysis, alerting, and dashboarding.
Ingested logs are all logs collected throughout your environment.
Ingestion control refers to the mechanisms and rules in the Agent and in tracing libraries for determining what traces are sent from an application to Datadog.
A Datadog default retention filter that is always active, keeping a representative proportion of traces, true high latency, and diverse error traces to help you monitor the health of your applications. It is not random, and so traces only retained by Intelligent Retention are not included in trace metrics.
IAST is a security testing methodology that combines static and dynamic testing.
In serverless computing, an invocation is when a deployed function is called.
Kubernetes is a platform for managing containers.
Synonyms: data link layer
In the OSI model of computer networking, layer 2 defines the network data format. Layer 2 concerns frames and physical addressing.
Synonyms: network layer
In the OSI model of computer networking, layer 3 determines how data is physically routed from source to destination. Layer 3 concerns packets and logical addressing.
Live Tail is all logs ingested by Datadog after processing but before indexing or archiving.
Log indexing filters logs into value groups for different retention periods, quotas, usage monitoring, and billing.
In Kubernetes, a manifest is a file that describes the creation and management of resources in a cluster.
A measure is a quantitative facet that can be used to aggregate values from multiple logs, filter logs using a range, or sort logs against a value.
Minified code is code (often JavaScript) stripped of comments, extra whitespace, unused code, and anything else that does not affect functionality. Minified code is less human-readable, but its smaller size improves website performance.
MITRE ATT&CK is a knowledge base of cyber adversary tactics and techniques.
A multi alert applies the alert to each source according to the monitor’s group parameter. An alert notification is sent for each group that meets the set conditions. For more information, see the documentation.
Multi-org is an account feature to manage multiple child organizations from one parent organization account. Users can be added to the parent-org and multiple child-orgs.
A multistep API test has several HTTP requests chained together to monitor user journeys on your services. For more information, see the documentation.
Mute a monitor to silence a monitor’s alerts and notifications.
NetFlow is a network protocol system that collects IP network traffic as it enters or exits an interface. It was introduced by Cisco in 1996.
Datadog’s Network Device Monitoring (NDM) provides visibility into on-premise and virtual network devices, such as routers, switches, and firewalls. For more information, see the documentation.
Datadog’s Network Performance Monitoring (NPM) provides visibility into network traffic between services, containers, availability zones, etc. For more information, see the documentation.
A network profile is set of attributes that describe how a network is configured.
No Data is when an integration or application is no longer submitting metrics to Datadog.
The Node Agent is the version of the Datadog Agent that runs on a host.
An OID is a standardized name used to identify an object.
OWASP is an organization that provides web application security resources.
The operational status of a port (up/down) refers to whether the port is up or down.
In a containerized infrastructure, an orchestrator automates the management of containers. This includes the provisioning, deploying, scaling, and networking of containers.
A parent org can manage and view the usage of multiple child organizations.
A pattern is when log messages have a similar structure. Pattern aggregation groups these logs together and displays them in the patterns view.
A pipeline is an ordered set of processers applied to a filtered subset of logs, which happens after the collection of logs but before indexing.
In Kubernetes, a Pod is the smallest deployable unit of computing.
In computer networks, PBR is a technique for routing data according to policies and filters.
In Datadog Synthetic Monitoring, a private location is a Docker container that a user can install inside a private network. This enables users to monitor internal-facing applications or private URLs that are not accessible from the public internet. For more information, see the documentation.
Related terms: processor , pipeline
For Datadog Events, a processing pipeline is a set sequence of data-structuring actions on event attributes when they are ingested. Users can configure processing pipelines to normalize and enrich events. For more information, see the documentation.
A processor is a set of instructions executed in a log pipeline to restructure log data and generate attributes to enrich logs.
A profile is snapshot in time of how much work (CPU usage, memory usage) is being done by code.
A query is composed of the metric name, the time aggregator, the space aggregator, and the scope. For more information, see the documentation.
Rate is a metric type that takes the count and divides it by the length of the time interval. For more information, see the documentation.
RUM is a UX monitoring technology that records user interactions with a website or application.
RED stands for “Rate, errors, and duration,” three key metrics for evaluating the performance of some code.
A reference table lists entities in Datadog like customer details, service names, and information, or IP addresses. The information is represented by a primary key and the associated metadata. For more information, see the documentation.
Rehydration is when archived logs are recalled back into Datadog.
Synonyms: indexing
Mechanisms and rules for determining what traces are retained for 15 day storage. Datadog retains a certain amount (Intelligent Retention) and users can create custom filters.
A role defines the account permissions for users. In Datadog, there are three default roles: Admin, Standard, and Read-only. For more information, see the documentation.
RBAC is a method to control read and write access to account assets based on roles that are granted permissions and assigned to users. For more information, see the documentation.
RASP is a security technology that detects and prevents attacks in real time.
In the Log Explorer, saved views keep track of different search queries, customized default visualizations, and a selected subset of facets. Saved views are shared across your organization.
The scope uses tag(s) to filter the query. For more information, see the documentation.
In Kubernetes, a Secret is an object that can be used to store sensitive data, such as passwords, tokens, and keys.
SIEM is a field in computer security that uses data from security events to support threat detection, security incident management, and compliance.
The Sensitive Data Scanner is a stream-based, pattern matching service to identify, tag, and optionally redact or hash sensitive data.
SSRF is a type of exploit where an attacker uses a server to access or manipulate information.
Serverless is a cloud development and execution model in which a cloud service provider handles server infrastructure.
Serverless insights are automatically-generated indicators (such as high memory usage, cold start, out of memory, etc.) that Datadog uses to identify and flag Lambda functions that are failing or performing poorly.
A service account is a non-human user that can be assigned a role and own application keys. For more information, see the documentation.
A service check monitors whether the status of a specific service is up or down. For more information, see the documentation.
An SLA is an explicit or implicit agreement between a client and service provider stipulating the client’s reliability expectations and the service provider’s consequences for not meeting them.
An SLO is a target percentage for application performance over a specific period of time. For more information, see the documentation.
In APM, the Service Map visualization provides an overview of your services and their health. It decomposes your application into its component services and draws observed dependencies between them.
In Datadog RUM, a session is a type of event. A user session begins when a user starts browsing the web application. It contains high-level information about the user, including their browser and device.
Session replay is a technique in UX testing that replays a user’s journey on a website or application.
SNMP is a protocol for collecting, organizing, and modifying information about managed devices on IP networks.
An SNMP MIB is a collection of definitions for a managed object’s properties (such as data types, access permissions, etc.) For more information, see the documentation.
SNMP Traps are notifications sent from an SNMP-enabled device to an SNMP manager. When a network device encounters unusual activity, such as a sudden state change on a piece of equipment, the device triggers an SNMP Trap event. For more information, see the documentation.
An SDK is a set of tools that enable developers to create applications for a specific technology, such as an operating system or a programming language.
A log source is where logs are collected and ingested into Datadog.
A source map is a file that maps minified JavaScript code to the original source.
Space aggregation splits a single metric into multiple timeseries by tags such as host, container, and region. There are four aggregation options: sum, min, max, and avg.
For more information, see the documentation.
A span is a logical unit of work in a distributed system for a given period. Multiple spans construct a trace.
A span ID is a numerical identifier generated by the tracing library for a span. Together with trace IDs, they are used to correlate traces and logs in Datadog.
A span tag is a tag that is applied to a [span][#span], in the form of a key-value pair, to correlate a request with other telemetry (or to filter it in searches).
A standard attribute is from a default set of attributes. These default attributes can be customized to create a naming convention for your organization.
Synonyms: static analysis
SAST is a security testing methodology that analyzes a program’s source code or binaries.
A template variable is an attribute used to customize and route monitor notifications based on the alert details. For more information, see the documentation.
Synonyms: rollup
Time aggregation is how Datadog combines data points into time buckets. There are five aggregation options: sum, min, max, avg, and count. For more information, see the documentation.
A trace tracks the time spent processing a request, and the status of this request. Each trace consists of one or more spans.
The trace ID is a numerical identifier generated by the tracing library for a trace. Together with span IDs, they are used to correlate traces and logs in Datadog.
A transaction aggregates indexed logs based on an instance of a sequence of events, such as a user session or a request processed across multiple micro-services.
A user is someone who has access to data in Datadog based on their assigned role.
In Datadog RUM, a view is a type of event. A view event is generated each time a user visits a web application page.
A warning is an optional monitor threshold setting for sending a warning notification, where the priority level is lower than an alert.
A WAF is a security tool that monitors and filters HTTP traffic from a web application.
A webhook uses a URL to connect your services and alerts your services when a metric alert is triggered. For more information, see the documentation.
