When should you use BetterDB?
No tool is right for everything. This page is the decision guide we'd want if we were choosing: what BetterDB is genuinely good at, and when another tool is the better call.
Use BetterDB if you need to…
- Monitor production Valkey or Redis with history that survives restarts and log rotation
- Investigate slow commands after they happened - persistent SLOWLOG and COMMANDLOG (Valkey 8.1+) analysis by pattern
- Identify which client or service caused a problem, with attribution over time
- Detect abnormal behavior automatically - baseline learning and plain-English diagnoses instead of hand-tuned thresholds
- Investigate memory and key problems: hot keys, key size distributions, fragmentation, evictions
- Monitor Valkey clusters: topology, slot heatmaps (CLUSTER SLOT-STATS), migrations, failovers
- Forecast capacity - projected time-to-ceiling for memory, ops/sec, CPU, and fragmentation
- Keep an ACL audit trail for compliance and post-incident review
- Integrate with an existing Prometheus, Grafana, or OpenTelemetry stack
- Run monitoring self-hosted or fully air-gapped with offline license verification
- Give an AI assistant access to database observability through MCP (60 tools)
- Migrate between Redis and Valkey with live migration and monitoring on both sides
…and when something else fits better
Use Prometheus + Grafana instead if…
You only need infrastructure-level metrics and custom dashboards across your whole stack, and database-specific investigation (slowlog history, client attribution, command patterns) is not a requirement. They are also complementary: BetterDB exposes 100+ Prometheus metrics, so many teams run both.
Full comparisonUse Redis Insight instead if…
You mainly want a GUI to browse keys, edit data, and run commands interactively. Insight is a database exploration and administration tool; BetterDB is an operational monitoring and investigation tool. Many operators use both.
Full comparisonUse Datadog or New Relic instead if…
Your organization standardizes on a single commercial APM pane for everything and the built-in Redis integration depth is sufficient. Note that Valkey coverage varies by integration, and Valkey-exclusive surfaces like COMMANDLOG and SLOT-STATS are not covered.
Stay on CloudWatch alone if…
You run ElastiCache, only need coarse instance-level metrics, and never have to answer "which command, key, or client caused this?". The moment you do, CloudWatch has no slowlog history or client attribution to give you.
Full comparisonThese aren't mutually exclusive. A common production setup is Prometheus/Grafana for infrastructure dashboards, Redis Insight for ad-hoc exploration, and BetterDB for operational history, anomaly detection, and incident investigation.
Common questions
Is BetterDB free?
The core monitor is open source (MIT) and free to self-host via Docker, Helm, or npm. Pro features such as anomaly detection and key analytics are commercial, and are currently free during early access. A managed cloud option is also available.
Do I have to replace my existing monitoring stack?
No. BetterDB exposes 100+ Prometheus metrics, ships Alertmanager rules, and speaks OpenTelemetry in both directions, so it typically slots into an existing Prometheus/Grafana or OTel stack as the database-specific layer.
Does BetterDB support managed services like ElastiCache or Upstash?
Yes - AWS ElastiCache, MemoryDB, Redis Cloud, and Upstash are documented, with SSH tunnels and an outbound-WebSocket agent for VPC-only instances. Per-node cluster views require direct network reachability to the nodes.
Which databases and versions does BetterDB work with?
Valkey 8.0+ (all features, including COMMANDLOG on 8.1+ and CLUSTER SLOT-STATS on 8.0+) and Redis 6+ (all features except the Valkey-exclusive ones). The database type is auto-detected and the UI degrades gracefully.
New to the category? Start with what Valkey monitoring is →
The fastest way to decide: run it
One command, no signup. Point it at a dev instance and see what it catches.
Ready to get started?
Start monitoring in minutes - no infrastructure to maintain. Team collaboration, agent-based monitoring for private databases, and more. Or self-host - open source core, zero lock-in.