Utilities

DevOps & SRE Tools

16 free, no-signup utilities for Site Reliability Engineers, DevOps teams, and platform engineers. SLOs, error budgets, burn-rate alerts, observability cost, Kubernetes right-sizing, postmortems and runbooks — all in your browser.

SLOs & Availability

Define realistic service-level objectives, measure how fast you are spending the error budget, and understand the availability ceiling imposed by your dependency graph.

Cost & Sizing

Quantify the dollar cost of downtime, estimate observability spend before signing the contract, and right-size Kubernetes pods so you stop paying for capacity you do not use.

Observability

Catch the high-cardinality label that is OOM-killing your Prometheus, lint your alert rules against best practices, and split a latency budget across every layer of your stack.

Incident Response

Measure the metrics that actually matter, generate the postmortem and runbook templates that on-call engineers actually want to read, and score your alerting hygiene before the next bad week.

Utilities

Everyday quality-of-life tools every SRE eventually wants on hand.

Automate your incident response

Reduce MTTR by 90% with AI-powered root cause analysis. Free to start.

Try Uptimes.ai Free

Built by SREs, for SREs

Every team running production systems eventually rebuilds the same utilities in a spreadsheet or scratch script: how much downtime does a 99.95% SLA actually allow each month? What did that two-hour outage cost the business? Is our 47-minute MTTR good or bad? Is this Prometheus rule going to flap? We built these tools so you can stop re-deriving the same answers and get back to the work that actually moves reliability forward.

Each tool runs entirely in your browser. There is no signup, no email gate, no logging of your inputs — you can audit the network tab and you will see nothing leaves the page. Link directly to any tool from your runbooks, onboarding docs, or SRE handbook.

When to use which tool

Reliability engineering tasks usually combine several of these tools. Here are the most common workflows we see in the wild.

Defining a new SLO from scratch

  1. Use the SLA & Uptime Calculator to translate business expectations into a concrete percentage.
  2. Use the Composite Availability Calculator to verify the target is achievable given your dependencies.
  3. Use the Burn Rate Calculator to derive the multi-window alert thresholds you will deploy.

Responding to a live incident

  1. Open the matching runbook (or generate one with the Runbook Generator if it does not exist).
  2. After resolution, run the MTTR Calculator with the incident data to see how you compare to last quarter.
  3. Generate a blameless postmortem with the Postmortem Template Generator.

Reducing observability cost

  1. Use the Datadog Cost Estimator to model the bill against current usage.
  2. Run the Prometheus Cardinality Estimator on your top metrics to find the high-cardinality offenders.
  3. Lint your rule files with the Prometheus Config Linter to catch flapping alerts before they ship.

Reclaiming an on-call rotation

  1. Run the Alert Noise Score to quantify current alerting hygiene.
  2. Use the On-Call Fairness Analyzer to spot rotation imbalance and burnout signals.
  3. Refine alert rules through the Burn Rate Calculator so on-call only fires on genuine SLO breaches.

SRE & DevOps Glossary

Terms and acronyms that show up across the tools above. Bookmark this section — many of these come up in every SLO review.

SLIService Level Indicator
A measurable signal of service quality — e.g., the fraction of HTTP requests that return successfully, or the fraction faster than 300ms.
SLOService Level Objective
A target for an SLI over a window — e.g., "99.9% of requests succeed over 30 days". The internal commitment your team makes.
SLAService Level Agreement
A contractually-binding promise to customers, usually with financial penalties (SLA credits) if you miss it. SLAs are typically set below SLOs to leave operational headroom.
Error budget
The maximum amount of unreliability allowed before violating the SLO. For 99.9% over 30 days, the error budget is 43 minutes 12 seconds of allowed downtime.
Burn rate
How fast you are consuming the error budget vs the ideal pace. 1× = on track, 2× = will exhaust budget in half the period, 14.4× = consuming 2% of monthly budget per hour.
MTTRMean Time to Recovery
Average time from incident start to full service restoration. The single most-tracked reliability metric.
MTTAMean Time to Acknowledge
Average time from alert firing to a human acknowledging it. Measures alert routing and on-call responsiveness.
MTTDMean Time to Detect
Average time from incident start to the team becoming aware (alert fires, customer report, manual observation).
MTBFMean Time Between Failures
Average time between successive incidents. Measures system reliability rather than response speed.
Cardinality
In Prometheus, the total number of unique time series for a metric. Calculated as the product of unique values across all labels — high cardinality is the #1 cause of Prometheus OOMs.
Recording rule
A precomputed PromQL expression saved as a metric. Used to make expensive aggregations instantly queryable. Convention: <level>:<metric>:<operation>.
Multi-window burn-rate alert
An alert that fires only when both a long window (e.g., 1h) and a short window (e.g., 5m) exceed a burn-rate threshold. Filters flapping while keeping detection fast.
QoS classQuality of Service
Kubernetes assigns each pod a class (Guaranteed, Burstable, BestEffort) based on requests/limits. Determines eviction priority under node pressure.
OOMKillOut-of-Memory Kill
When a process exceeds its memory limit, the kernel terminates it. The most common preventable Kubernetes failure mode.
eBPFExtended Berkeley Packet Filter
A kernel feature that lets safe programs run inside the Linux kernel. Used for zero-instrumentation network observability — see TCP latency, retransmits, and connection errors without touching application code.
Blameless postmortem
An incident review focused on systems and processes, not individuals. Premise: people act in good faith with the information they have, so "person X did Y" is never the root cause.
5 Whys
A root-cause technique: keep asking "why did that happen?" until you reach a systemic cause. Most teams stop at the second why and miss the real driver.
Runbook
A short, action-oriented document for on-call: what the alert means, how to confirm, how to mitigate. Linked from the alert itself for 3-AM reachability.
Toil
Manual, repetitive, automatable work that scales linearly with the service. Google's SRE book defines it precisely; the cure is usually code.
Flapping alert
An alert that repeatedly fires and resolves on its own without action. Almost always means a missing for: clause or a too-tight threshold.
Page
An alert urgent enough to wake someone up. The opposite of a ticket. Rule of thumb: only page if doing nothing makes the situation materially worse.
Canary deploy
Releasing a new version to a small fraction of traffic first to detect regressions before full rollout. The most cost-effective deploy safety mechanism.

From utilities to automation

Utilities like these describe the problem. Reducing MTTR — actually moving from 91 minutes to 47 seconds — takes automation. Uptimes.ai is an AI SRE agent that lives inside your Kubernetes cluster, correlates alerts across Datadog, Prometheus, and PagerDuty, reasons over your service dependency graph using eBPF kernel-level visibility, and produces a root cause report in under three minutes. When automation can resolve the issue — common cases like CrashLoopBackOff, OOMKilled, deploy regressions — it does, with a full audit trail.

If your numbers from the MTTR Calculator are higher than you would like, that gap is exactly what we exist to close. See how Uptimes.ai works →

Frequently Asked Questions

Are these tools really free?+
Yes, every tool on this page is 100% free, with no signup, no email gate, and no usage limits. They run entirely in your browser — your inputs never leave your device.
Which tool should I use first?+
If you are setting or reviewing an SLA, start with the SLA & Uptime Calculator. If you are tracking SLO burn this period, the Error Budget Calculator. If you are sizing alert thresholds, the Burn Rate Calculator. If you are building a business case for reliability work, the Incident Cost Calculator. Use the workflow guide above to find the right tool for your current task.
Why did Uptimes.ai build these tools?+
We talk to SRE and platform teams every week, and these are the calculations and lookups they reach for most often — usually re-implementing them in spreadsheets or scratch scripts. Building them once, correctly, and giving them away makes our community stronger and gives us a chance to introduce people to Uptimes.ai's automated incident response platform.
Do you store anything I enter?+
No. Every tool is fully client-side. We do not log inputs, results, or anything beyond standard anonymous traffic analytics on the page itself.
Can I link to or embed these tools?+
Absolutely — link directly to any tool page. We are happy when SRE handbooks, runbooks, and onboarding docs point to these utilities. Attribution is appreciated but not required.
Will you build more tools?+
Yes — we regularly add new tools based on what teams reach out about. If there is a calculation or template you keep rebuilding in a spreadsheet, tell us via the Contact link and we will probably build it.
How is this different from the SRE handbook chapters?+
Google's SRE Workbook (free online) explains the concepts; these tools execute the math. Use both: read the chapter to understand burn-rate alerts, use the Burn Rate Calculator to generate the actual thresholds you will deploy. The tools also include best-practice opinions baked in (recommended QoS classes, sensible default windows, etc.) where the books leave it to interpretation.