Listening: Enterprise SLA Review Meeting

SLA (Service Level Agreement) is a contract between a service provider and a customer that defines the level of service expected — typically including uptime guarantees, response times, and remedies for failures.

The mathematics of 99.9% uptime:

• 99.9% uptime = ~8.7 hours of permitted downtime per year
• ~43.8 minutes per month
• 4 hours of downtime in a quarter = approximately 4× the monthly allowance — a clear breach

Key SLA vocabulary:

• uptime — percentage of time the service is available
• downtime — period when the service is unavailable
• SLA breach — the vendor failed to meet the contracted level of service
• remediation / credits — compensation provided when an SLA is breached (often service credits)

The SLA / SLO / SLI framework:

• SLA (Service Level Agreement) — the external contract with the customer. Example: "We guarantee 99.9% uptime."
• SLO (Service Level Objective) — the internal target the engineering team sets. Often stricter than the SLA. Example: "We aim for 99.95% uptime internally."
• SLI (Service Level Indicator) — the actual measured metric. Example: "Last month's measured uptime was 99.7%."

Relationship: SLI is what you measure. SLO is what you aim for. SLA is what you promise. If your SLI falls below your SLO, you may breach your SLA.

Why the distinction matters in a support meeting: Reviewing SLIs against SLOs allows the vendor to show the customer what was measured versus what was targeted — providing evidence and context for the incident discussion.

Blast radius is borrowed from military terminology and used in IT to describe the scope of impact of a failure — how many users, services, or regions were affected.

Usage in incident language:

• "The blast radius was limited to a single availability zone" = only one zone was affected
• "We need to reduce the blast radius of future deployments" = deploy gradually to limit impact if something goes wrong

Related incident vocabulary in this sentence:

• circuit breaker — a pattern that stops calls to a failing service to prevent cascading failures
• cascading failure — a failure in one component that causes other components to fail in sequence
• subset — a portion of customers (not all); contrast with "all customers"

Professional note: "Blast radius" is widely used in SRE (Site Reliability Engineering) contexts. Hearing it in an enterprise support meeting signals that the speaker has engineering background.