Elastic’s new metrics capabilities will dramatically improve uptime for public sector IT

Many public sector organizations already rely on Elastic for affordable log storage and retention, especially to comply with government mandates, such as OMB Memorandum M-26-14 in the US.

Now, that same trusted logging foundation extends to metrics. Most public sector organizations running modern infrastructure have accumulated observability debt in the form of siloed tools: one system for logs, another for metrics, and a third for traces. Every signal lives in a different back end, uses a different query language, and requires context-switching to correlate. When an incident occurs, engineers are assembling a picture by hand across disconnected systems under pressure.

Elastic Observability stores metrics, logs, and traces in a single unified back end. When an alert fires, the investigation context is already assembled in the form of correlated log events, preceding metric anomaly, or related trace. Engineers aren't opening three tabs before they can form a hypothesis. Machine learning (ML) anomaly detection runs automatically against infrastructure metrics, so alerts surface with context about what changed and how severe the deviation is, not just a raw threshold breach. For public sector SREs managing services that constituents depend on, this means faster mean time to resolution. For leadership, it means fewer prolonged outages and less personnel time spent on manual correlation.

Elastic Observability runs across three deployment modes: Elastic Cloud Serverless, Elastic Cloud Hosted, and fully self-managed on-premises or air-gapped environments. Unlike some vendors that limit on-premises deployment or reserve their highest-value features for hosted instances, Elastic's full capabilities are available across all three. Organizations with strict data residency requirements can run the complete platform in their own infrastructure without sacrificing capability.

Public sector IT budgeting depends on predictability. Multiyear contracts, budget cycles, and appropriations processes don't accommodate surprise invoices driven by opaque pricing mechanics.

Many commercial observability vendors use pricing models that compound unpredictably as modern infrastructure grows. Per-host fees are combined with custom metric charges and container overages, creating bills that climb as Kubernetes environments scale or OpenTelemetry instrumentation deepens. The costs are hard to forecast because they're tied to cardinality — the very property that makes observability valuable in complex environments.

Elastic's model is fundamentally different: You pay for data volume, not for how many hosts you have or how granularly you instrument your systems. On Elastic Observability Serverless, metrics stored in time series data streams (TSDS) index mode are priced at 25% of the standard Observability per-GB rate for both ingest and retention, or roughly $0.023 per GB ingested and $0.005 per GB per month retained at volume tiers. There are no per-host licenses, no cardinality-based surcharges, and no custom metric classifications that penalize richer instrumentation, landing at half the cost of Datadog and other vendors.

Long-term retention is also meaningfully affordable; at roughly $0.005 per GB per month, keeping a year or more of full-resolution metrics data is financially viable, supporting the kind of retrospective auditing that public programs often require.

Public sector infrastructure evolves slowly and deliberately. Procurement cycles are long, integrations run deep, and the operational risk of disruption to constituent-facing services is high. Any platform transition has to protect what's already been built.

Most SRE teams running modern cloud and containerized infrastructure have invested significantly in Prometheus-based tooling: scrape configs, alert rules, PromQL queries, and Grafana dashboards that represent years of operational knowledge. Historically, migrating metrics back ends meant rewriting all of it.

Elastic has removed most of that friction. Prometheus metrics arrive via Prometheus Remote Write and land in the same columnar store without semantic changes. PromQL works natively in Kibana, so existing queries, dashboards, and alert rules migrate directly without modification. For organizations that want to keep Grafana as their visualization layer, Elasticsearch exposes a native Prometheus-compatible API that any compatible front end can query directly, meaning teams can replace the back end while keeping the interface their engineers already know.

Migration can be incremental and largely automated. The Observability Migration Platform is a CLI-driven workflow that translates supported Grafana and Datadog assets into Kibana-native outputs and produces the evidence needed to review the result, changing migration from a manual rebuild into a translation-and-verification workflow. It works from exported assets or live APIs, covering dashboards and alerting content from both Datadog and Grafana paths. Crucially for public sector teams, the workflow also produces reviewer-facing evidence, such as a migration report, manifest, verification packets, and rollout plan, so teams can see what translated cleanly, what was downgraded or flagged for manual review, and what still needs human judgment.

Elastic Observability’s credibility in the market continues to grow. In July 2026, Elastic was named a Leader in the 2026 Gartner® Magic Quadrant™ for Observability Platforms for the third consecutive year. We believe this recognition reflects Elastic’s prioritization of staying ahead of market shifts while offering customers the efficiencies of standardization through Prometheus and OTel as well as the ability to use a single platform for logs, metrics, and traces.

Learn more about how public sector leaders are approaching observability by downloading the 2026 Landscape of Observability for Public Sector report, or schedule time with one of our experts.