[BLCCoreStudio]

Summary

Upgrades cloud/prometheus from a compact schema-v1 overview into a schema-v3 deep technical reference organized around the real metrics lifecycle: discovery, relabeling, scraping, ingestion, local TSDB storage, PromQL, rule evaluation, alerting, federation, and remote-storage integration.

Changes

• explain Prometheus server, exporters, client libraries, Pushgateway, Alertmanager, and API/visualization boundaries
• document the labeled time-series model, counters, gauges, classic/native histograms, summaries, staleness, and series identity
• cover pull-oriented scraping, service discovery, target/metric/write relabeling, and Kubernetes discovery boundaries
• explain TSDB head, WAL, blocks, compaction, retention, filesystem assumptions, backup/restore, and disk headroom
• add PromQL evaluation types, rate/reset semantics, aggregation, vector matching, recording rules, alerting rules, and common query mistakes
• treat cardinality and churn as first-class capacity and reliability concerns
• cover remote write/read, federation, HA replica patterns, long-term/distributed-storage boundaries, and scaling levers
• add Kubernetes, Grafana, and OpenTelemetry integration boundaries
• add security, failure modes, debugging workflows, alert design, anti-patterns, capacity planning, and a staged learning path
• add Technology Universe coverage metadata and typed relationships
• verify Apache-2.0 and expand primary sources to current upstream data-model, query, rule, storage, remote-write, federation, instrumentation, and API documentation

Quality bar

This follows the project's deep-dive standard without turning into a PromQL command catalog. The organizing model is the end-to-end metrics pipeline plus the cardinality/storage/query economics operators need to reason about real Prometheus failures.