AI Tools

Velero backs up Kubernetes resources and persistent volumes to cloud storage (S3, GCS, Azure). Disaster recovery: restore to new cluster in minutes. Migration tool for multi-cluster ops. Hooks for databases (e.g., freeze during backup). Open-source, used by Shopify.

Weave provides encrypted networking for Kubernetes pods with automatic routing. Mesh topology auto-discovers new nodes. Policy engine controls traffic between pods. Integration with Prometheus for observability. Low memory footprint. Used in production at scale.

Bezel profiles Java applications in production to identify CPU and memory bottlenecks without code redeployment. The profiler traces lock contention and garbage collection pauses with nanosecond precision. Historical data storage enables comparison of performance across releases. Integrates seamlessly with existing APM agents and monitoring dashboards. Engineering teams founded by distributed systems veterans from major tech companies.

Coralogix is an observability platform combining metrics, logs, and traces with telemetry pipeline to normalize costs. Machine learning flag anomalies. Supports any log format. API-first for custom integration. Series-B, growing among Israeli and European enterprises.

Datadog creates synthetic tests for APIs and web applications. Record user journeys. Continuous validation from multiple regions. Integrates with Datadog monitoring. Unified observability platform.

Dynatrace uses AI to automate root-cause analysis of performance issues across applications, infrastructure, and networks. Detects anomalies and suggests fixes without manual thresholds. Processes metric, log, and trace data from 1M+ monitored entities. Auto-discovers application topology. Series-D company, trusted by enterprises like Samsung and NBC.

Elasticsearch 8+ supports dense vectors and ANN search. Integrates with existing Elasticsearch clusters. Combine dense and sparse retrieval. Vector store for LLM retrieval. Widely adopted.

Honeycomb is the observability platform purpose-built for complex microservices. Send OpenTelemetry traces. Honeycomb auto-indexes all fields. Ask questions in plain English (What was the p99 latency for checkout yesterday?). BubbleUp highlights the top correlations with latency spikes. Series-D, used by Figma and Twitter.

InfluxDB is optimized for metrics and events at high cardinality. Downsampling reduces long-term storage. Continuous aggregates compute sums pre-emptively. InfluxQL and Flux query languages. Cloud and self-hosted. Used by Tesla and Cisco.

Jaeger is an open-source platform for tracing microservice architectures. Understand request paths across 20+ services. Supports sampling to reduce storage load. Backend options: Cassandra, Elasticsearch, Badger. UI shows service topology and critical paths. Used by Uber, PayPal, and DoorDash. Graduated CNCF project.

Logz.io provides Elasticsearch, Logstash, and Kibana as a managed service. Ingest logs from servers, containers, apps. Machine learning detects anomalies. Compliance reports auto-generate for SOC2. SIEM adds security analysis. Series-D, trusted by 10,000+ teams.

OpenObserve is an open-source log platform optimized for cost. Parquet storage and columnar compression cut costs vs Splunk by 80%. Single API for logs, metrics, and traces. Sub-second query latency. Built-in ingestion of syslog, JSON, CEF formats. Series-A company.

OpenTelemetry is a vendor-neutral standard for collecting metrics, traces, and logs from any application. Collector receives data from SDKs, transforms, and exports to backends (Datadog, Grafana, Splunk). No lock-in: swap backends anytime. Specification published by CNCF. De-facto standard in observability.

Quickwit is an open-source search engine for logs optimized for cost. Parquet columnar format and streaming storage compress 100x. Sub-second full-text search on terabytes. Native OpenTelemetry support. Built by ex-Datadog engineers. Ideal for teams with massive log volume.

Rollbar tracks exceptions and errors in production, grouping by pattern. Integrates with CI/CD to show which release introduced a bug. Notifies engineers and creates tickets. Version history shows when errors started. Used by Shopify and Twitch. Series-C company.

Scout APM instruments Node.js, Python, and Ruby apps with minimal overhead (1-3% CPU impact). Automatic request profiling pinpoints slow code. Session replay captures user actions during errors. Slack integration alerts on anomalies. Bootstrapped company focused on developer experience. Ideal for early-stage startups.

Sensu is an event-driven monitoring and alerting platform for hybrid infrastructure. Agents collect metrics and check status. Central Sensu handler routes alerts to Slack, Kafka, or custom webhooks. Built-in deduplication prevents alert storms. Multi-region replication for high availability. Popular with ops teams.

Splunk Enterprise searches events and logs with SPL language. Indexes everything for fast ad-hoc queries. Time-based analysis. Compliance reports. Market leader for enterprise search. IPO company.

TraceHarbor APM instruments applications in 20+ languages (Python, Go, Java, Node, .NET) to trace requests end-to-end. Tag traces with deployment version and custom attributes. Smart sampling retains rare-but-important errors while shedding normal traffic. Integrates with ServiceNow and PagerDuty for incident response. Market leader with 18,000+ customers.

Apache NiFi routes data between systems with visual dataflow composition and no code. Built-in backpressure prevents pipeline bottlenecks. NiFi's guaranteed delivery, flow-level lineage, and 200+ processors cover JSON parsing, geolocation, regex, and 3rd-party API calls. Security includes LDAP, Kerberos, and data masking. Large banks and telecom firms run NiFi on thousands of nodes.