Opentelemetry Collector

Spotify's Annoy library indexes high-dimensional vectors in memory. Fast search and low memory usage. Python and C++ implementations. Used internally by Spotify. Active maintenance.

APISIX is an open-source cloud-native API gateway. Dynamic routing and plugin loading. Multi-protocol support (HTTP, gRPC, Dubbo, WebSocket). Metrics exported to Prometheus. Helm chart for Kubernetes.…

Apache Cassandra stores time-series at petabyte scale. Write-heavy workload optimized. Time-bucketing for efficient queries. Replication across regions. Used by Apple and Netflix.

Chroma is an open-source embedding database built for AI applications. Run locally or distributed. SQLite backend. Hugging Face integration. Simple API. Easy to get started.

eBPF programs run safely in the Linux kernel. Monitor system calls, network, disk. No recompile needed. Used by Cilium, Falco, and Pixie for observability. New programming model transforming Linux inf…

Meta's Faiss library searches billions of vectors. GPU acceleration with CUDA. Index compression reduces memory. Research and production use. Widely used in recommendation systems.

Freebase is a large collaborative knowledge base. Now part of Google Knowledge Graph. CC licensed, freely available for download. Groundtruth for knowledge graph research. 43M+ entities extracted.

GQL is a standardized graph query language (ISO/IEC 39075). Successor to Cypher. Multi-vendor support planned. Early adoption by industry players. Will unify graph query ecosystem.

Gradio creates simple interfaces for ML models. Share models via public link. Input/output components. LaunchPad for easy deployment. Hugging Face integration.

Grafana Loki is a horizontally scalable log aggregation system. Label-based indexing stores logs cost-effectively. LogQL queries filter by service, pod, region. No high cardinality concerns. Pairs wit…

Helm packages Kubernetes applications as charts, bundling manifests, values, and dependencies. Render environment-specific values (dev, prod) from one chart. Rollback previous releases with one comman…

Hex is a notebook environment for data analytics teams that bridges Jupyter and Dashboards. Write SQL, Python, and R in reactive cells. Parameters auto-build filters without code. Share notebooks as i…

KubeVirt lets you run virtual machines on Kubernetes like pods. Useful for legacy VMs or Windows workloads. Networking and storage APIs consistent. Live migration. Operated as a DaemonSet. Community p…

Marqo is an open-source tensor search engine. No API calls to embeddings service. Local document indexing. Query-specific fine-tuning. Built for ease of use.

Metricbeat from Elastic ships metrics to Elasticsearch. Modules for common services (Docker, Postgres, Redis). Lightweight agent. Integrates with Kibana visualizations. Part of Elastic Stack.

Milvus is an open-source vector database for large-scale similarity search. Billion-vector scale. Multiple index types: IVF, HNSW, DiskANN. Cloud-hosted or self-hosted. Supports multiple languages. CN…

Desktop app for running and chatting with local AI models with RAG, web search, and model management.

NGINX Ingress Controller exposes Kubernetes services via HTTP/HTTPS. Terraform and Helm deploy easily. Rewrite and regex support. SSL/TLS termination. WebSocket passthrough. High performance at scale.…

NMSLIB provides approximate nearest neighbor search. C++, Python, Java, Ruby bindings. HNSW and other algorithms. High performance tuning options. Research origins.

OpenTSDB stores time-series on top of HBase. Billions of metrics at millisecond precision. Tag-based queries. Built-in aggregators for rollups. Java-based backend.

Self-hosted web interface for interacting with local and remote language models through a familiar ChatGPT-style chat UI. Supports Ollama, OpenAI API, and other backends. Features include RAG for quer…

One-click installer for AI applications that sets up Stable Diffusion, LLMs, and other AI tools locally.

Ponchao is Tencent's open-source chaos testing framework. Multi-platform support (cloud, on-premise). Orchestrates complex scenarios. Real-time status monitoring. Growing adoption in Asia.

pgvector is an open-source extension for Postgres. Store and search vectors in Postgres. Index types: IVF, HNSW. No separate database needed. Simple to deploy. Community-maintained.

Open source tool for chatting with your documents privately using local LLMs without data leaving your machine.

Prometheus scrapes metrics from HTTP endpoints every 15 seconds. Time-series with labels enable multi-dimensional queries. Pull-based avoids overwhelming servers. AlertManager routes incidents. De-fac…

Qdrant is an open-source vector database optimized for semantic search and recommendation systems. HNSW indexing with pruning. Payload storage with filtering. Snapshots and recovery. Rust implementati…

SigNoz is an open-source alternative to Datadog combining metrics, traces, and logs. Stores data in ClickHouse for cost efficiency. Alerts integrate with Slack, PagerDuty, and Webhook. Self-hosted or …

Open source chat interface for interacting with AI models with character cards, world building, and extensions.

StatsD is a lightweight protocol and reference implementation for publishing application metrics. Applications send counters, timers, and gauge values via UDP packets to a local agent. The agent aggre…

Streamlit rapidly builds ML web apps in Python. Interactive widgets with no frontend coding. Real-time reruns on code changes. Deployment to Streamlit Cloud. Developer-friendly.

Telegraf is a plugin-driven server agent for collecting metrics. 200+ input plugins (CPU, disk, Docker, Prometheus). Output to InfluxDB, Graphite, or Kafka. Lightweight, single binary. Standard in mon…

Thanos is a set of components extending Prometheus. Sidecar uploads blocks to S3. Querier aggregates across all Prometheus instances. 5-year retention. Ruler for alert generation. CNCF project.

Vald is an open-source distributed vector database. High-dimensional approximate nearest neighbor search. Horizontally scalable. Python and Go clients. Japanese origin, growing adoption.

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 dat…