Dynatrace Professional flashcards
Dynatrace Professional · Build recall with a focused review session.
32 cards. Read the term and the definition here. The deck at the top of this page is optional.
- Davis causation vs. prediction
- Causation is deterministic root-cause analysis over the Smartscape topology. Prediction is probabilistic anomaly detection from adaptive baselines. An answer should state which engine produced it.
- Adaptive baseline
- A learned per-metric model that follows daily and weekly seasonality. A violation is a deviation from this baseline, not a fixed numeric threshold.
- Predictive alerting
- Extrapolates a metric trend to forecast a threshold breach before it happens. It complements static alerting profiles rather than replacing them.
- Davis CoPilot
- An assistant layer over Grail and Davis that generates DQL queries and explanations. Its outputs must be validated against the underlying data before acting.
- Service definition
- A logical grouping of process groups exposing the same endpoint. It is the primary unit of analysis, and requests are traced across service boundaries.
- PurePath vs. distributed trace
- PurePath captures one trace with method-level timing. Distributed tracing stitches spans across services and processes into a single end-to-end view.
- Service-level objective
- A target such as 99.9% of requests under 500 ms. The error budget is the permitted shortfall, and burn rate measures how fast it is consumed.
- Request attributes
- Dimensions such as HTTP method, URI, or custom attributes that slice response time and failure rate without new instrumentation.
- Runtime vulnerability analytics
- Evaluates third-party libraries actually loaded at runtime. A vulnerable dependency that is never loaded is not reported as an active risk.
- Runtime application protection
- Detects attacks such as SQL injection, command injection, and SSRF at process level. It can block them in-line without a separate web application firewall.
- Security risk scoring
- Davis risk assessment combines vulnerability severity (CVSS), exposure (public vs. internal), and the sensitivity of data assets the process touches.
- Vulnerability lifecycle states
- Open, resolved, and muted. Muting requires a documented justification and a review date so suppressed findings do not become permanent blind spots.
- Business events
- Grail records carrying business context such as order ID, customer, and amount alongside technical identifiers. They are queried with DQL and are not tied to host lifetime.
- Business flow
- A user journey linked across services where each step is a business event. Its conversion rate is the fraction of journeys completing all steps.
- Business metric derivation
- Business metrics are aggregated from business events, for example order count per minute or revenue sum. They can feed SLOs, dashboards, and alerting.
- Bizevent retention cost
- Business events have their own bucket retention. High-cardinality attributes raise storage and query cost, so attribute cardinality should be bounded.
- RUM vs. synthetic
- Real User Monitoring captures actual sessions via JavaScript beacons. Synthetic monitoring runs scripted checks from controlled locations on a schedule, independent of real traffic.
- Core Web Vitals
- LCP measures loading, INP measures responsiveness and replaced FID, and CLS measures visual stability. These are the standard user-experience performance metrics.
- Apdex formula
- (satisfied + tolerating/2) / total. It is a user-satisfaction index with configurable thresholds per application.
- Session replay privacy
- Session replay reconstructs the user interaction sequence. It is privacy-sensitive and requires masking of personal data before capture, not after.
- OneAgent deployment modes
- Full-stack, infrastructure-only, and discovery. The chosen mode determines which host, process, and network signals are collected for observability.
- Process group vs. instance
- A process group groups identical processes across hosts. A process group instance is the individual process running on a specific host.
- Smartscape topology
- A continuously refreshed model of hosts, processes, services, and their dependencies. It is the basis for root-cause traversal during incident analysis.
- Maintenance windows
- Suppress alerting during planned changes. They should be scoped to the affected entities and time range to avoid hiding unrelated incidents.
- DQL pipeline
- Log queries follow fetch, filter, summarize, producing a tabular record set. Projection operators such as fieldsAdd and fieldsRemove shape the output.
- Log processing rules
- Extract attributes such as severity and service from raw lines. Extracted attributes become queryable dimensions and enable log-to-trace correlation.
- Log pattern detection
- Groups similar log lines into a pattern. Anomaly detection on log volume or error rate can then trigger problems automatically.
- Log redaction timing
- Sensitive data must be redacted at ingestion. Once stored in Grail, redaction requires reprocessing and is not retroactive for already-ingested records.
- Release monitoring
- Tracks a deployment across environments. Release health is measured by error rate and response time against the pre-deployment baseline.
- Deployment event correlation
- Deployment events from CI/CD, Kubernetes, or the events API are correlated with service behavior to attribute a regression to a specific release.
- Automatic release validation
- Compares key metrics over a defined window after deployment. A failed validation should trigger rollback automation rather than manual inspection.
- Threat observability
- Continuous visibility of attack surface, exposure, and attacker behavior across runtime, correlating detection signals with vulnerability context to produce a threat score.