Skip to main content

Glossary

Key terms and concepts used throughout the KredSLA platform and documentation.


Cloud & Infrastructure

TermDefinition
SLA (Service Level Agreement)A contractual commitment from a cloud provider guaranteeing a minimum level of service availability, typically expressed as a monthly uptime percentage (e.g., 99.99%).
SLO (Service Level Objective)An internal performance target set by an organization, often more stringent than the provider's SLA.
SLI (Service Level Indicator)A quantitative measure of a specific aspect of service performance, such as uptime, latency, or error rate.
SLA CreditMonetary compensation (typically bill credits) owed by a cloud provider when their service fails to meet the agreed SLA.
Uptime PercentageThe fraction of time a service was available during a billing period. Usually measured monthly. Example: 99.95% uptime = ~21.9 minutes of downtime per month.
Multi-AZA deployment architecture spanning multiple Availability Zones within a cloud region for high availability.
Cross-Region ReplicaA database or service copy in a different geographic region for disaster recovery.
IAM RoleAn identity in AWS (or equivalent in other clouds) with specific permissions. KredSLA uses read-only IAM roles to access your cloud environment.
Service PrincipalThe Azure equivalent of an IAM role — an identity used by applications to access resources.

KredSLA Platform

TermDefinition
Cloud AccountA connected AWS, Azure, GCP, or OCI account that KredSLA monitors for SLA breaches.
Monitored ResourceA cloud resource (e.g., EC2 instance, RDS database, Cloud SQL instance) covered by an SLA and actively monitored for violations.
Discovery ScanThe automated process of enumerating all SLA-eligible resources in a connected cloud account.
Service Dependency GraphA tree structure mapping the relationships between cloud resources, used to trace outages to their root managed-service failure.
Evidence BundleA structured package of metrics, incident records, and SLA references assembled to support a credit claim.
Claim LifecycleThe progression of an SLA credit claim: Detected → Bundled → Filed → Approved/Rejected → Reconciled.
Observability BridgeThe KredSLA component that connects to cloud-native monitoring platforms (CloudWatch, Azure Monitor, GCP Monitoring, OCI Observability) for richer SLA detection.
SLA LibraryA read-only reference of cloud provider SLA definitions, including uptime thresholds, credit tiers, and filing windows.
Contingency FeeThe percentage of recovered credits that KredSLA charges as its service fee (default 20%). No recovery = no fee.
Savings WidgetThe dashboard panel displaying total recovered credits, trends, and provider-level breakdowns.
Claims TrackerThe dashboard table showing every SLA claim with its current status, evidence, and financial impact.

Claim Statuses

StatusDefinition
DetectedAn SLA breach has been identified and verified against the provider's official health feed.
BundledEvidence (metrics, incident records, dependency path, SLA reference) has been compiled for filing.
FiledA support case has been submitted to the cloud provider via their API.
ApprovedThe cloud provider has accepted the claim and issued a credit.
ReconciledThe credit has been confirmed on the customer's billing statement.
RejectedThe cloud provider denied the claim.

Security & Compliance

TermDefinition
SOC 2A compliance framework that evaluates an organization's controls for security, availability, processing integrity, confidentiality, and privacy. Type I assesses control design; Type II tests effectiveness over time.
GDPRThe General Data Protection Regulation — EU legislation governing how personal data is collected, processed, stored, and deleted.
Right to Erasure (GDPR Art. 17)A data subject's right to have their personal data deleted. KredSLA implements this as an immediate cascading deletion across all data stores.
DPA (Data Processing Agreement)A contract between a data controller (customer) and data processor (KredSLA) required under GDPR Article 28.
OpenBaoAn open-source fork of HashiCorp Vault used by KredSLA for secrets management. Cloud credentials are stored here, never in the application database.
RBAC (Role-Based Access Control)Access control based on user roles. In KredSLA, every API request is scoped to the caller's organization, preventing cross-tenant access.
JWT (JSON Web Token)A compact token format used for API authentication. KredSLA JWTs include issuer and audience claims with a 1-hour TTL.
2FA (Two-Factor Authentication)An additional verification step beyond username/password. KredSLA uses email-delivered one-time codes.
SAST (Static Application Security Testing)Automated code analysis for security vulnerabilities. KredSLA uses Bandit for Python SAST in CI/CD.
RPO (Recovery Point Objective)The maximum acceptable amount of data loss measured in time. KredSLA targets ≤ 5 minutes via RDS Point-in-Time Recovery.
RTO (Recovery Time Objective)The maximum acceptable time to restore service after a failure. KredSLA targets ≤ 1 hour.

Cloud Provider Terms

TermDefinition
AWS Health APIAmazon's API for programmatic access to service events and outage information.
Azure Service HealthMicrosoft's service for tracking the health of Azure resources.
GCP StatusGoogle's dashboard and API for reporting GCP service availability.
OCI StatusOracle Cloud's service health reporting.
CloudWatchAWS's monitoring service for metrics, logs, and alarms.
Azure MonitorMicrosoft's full-stack monitoring service for Azure resources.
GCP Cloud MonitoringGoogle's infrastructure monitoring service (formerly Stackdriver).
Filing WindowThe time period after an incident during which a credit claim must be submitted. Varies by provider: AWS (2 billing cycles), Azure (2 months), GCP (30 days).

FinOps

TermDefinition
FinOpsA financial operations practice for managing cloud costs, combining technology, finance, and business disciplines.
Cost AssuranceThe practice of ensuring organizations pay only for delivered performance — recovering credits when providers fail to meet SLA commitments.
Financial LeakageMoney lost through unclaimed SLA credits, undetected billing errors, or unoptimized resource usage.
Vendor AccountabilityUsing performance data and recovery history to hold cloud providers accountable during contract renewals.