Open-source alternatives to Petl

Apache Beam is a distributed data pipeline framework and unified data processing model designed to handle both bounded batch data and unbounded real-time streams. It provides a system for building scalable, data-parallel workflows that operate across compute clusters using a single programming model. The framework utilizes a cross-runner pipeline abstraction that decouples the data processing logic from the underlying execution backend, allowing the same pipeline to run on different distributed compute engines. It supports multi-language pipeline development by translating high-level code fro

Apache Flume is a distributed, reliable, and available service for efficiently collecting, aggregating, and moving large amounts of log-like data

Apache Pulsar is a cloud-native message queue and distributed publish-subscribe messaging system. It serves as a multi-tenant event streaming platform designed to route data streams for asynchronous communication between producers and consumers. The system distinguishes itself through geo-replication, synchronizing data across multiple geographic regions to ensure high availability and low latency. It implements a multi-tenant architecture that provides isolation and resource management for millions of independent topics. The platform covers high-throughput data streaming and event-driven da

Kafka is a distributed event streaming platform designed for capturing, storing, and processing real-time data streams across interconnected nodes. It functions as a distributed commit log, providing a fault-tolerant storage mechanism that records state changes sequentially to ensure data consistency and durability across distributed environments. The platform distinguishes itself through a partitioned commit log architecture that enables horizontal scaling and parallel processing of data streams. It integrates a stream processing engine for continuous transformations and aggregations, while

Apache NiFi is a flow-based programming platform that enables the visual design, monitoring, and management of data pipelines. At its core, it provides a web-based visual dataflow designer where users build directed graphs of processors to route, transform, and mediate data movement between any source and destination without writing custom code. The system records fine-grained data provenance for every data item from ingestion to delivery, supporting audit, debugging, and replay of data lineage. The platform distinguishes itself through a zero-master cluster architecture that distributes proc

Mirror of Apache Oozie

RocketMQ is a cloud-native distributed messaging platform and streaming engine. It functions as a distributed transactional queue that ensures atomicity between local transactions and message delivery, and serves as an MQTT IoT message broker to bridge lightweight device traffic into high-performance data streams. The system is distinguished by a Kubernetes-native architecture that decouples compute from storage to allow independent scaling of traffic and data retention. It utilizes a tiered storage model to offload older data to remote storage and employs quorum-based replication and automat

Argo Workflows is a container-native workflow engine that functions as a Kubernetes custom resource controller. It orchestrates complex sequences of containerized tasks by executing them as directed acyclic graphs, allowing for dependency management and parallel processing within a cluster. The system extends the native Kubernetes control plane to manage the full lifecycle of automated processes, from initial triggering to final resource cleanup. The platform distinguishes itself through its controller-pattern reconciliation, which continuously monitors workflow states to align them with desi

Arrow is a Python date and time library that provides a simplified interface for creating and manipulating timestamps by wrapping the Python standard library. It serves as a tool for managing date objects, handling timezone offsets, and performing relative date calculations. The library is distinguished by its ability to humanize timestamps into natural language relative descriptions across multiple locales and parse human-readable time phrases back into precise date objects. It also features a specialized parser for converting ISO 8601 strings and custom formatted text into date objects. Br

AutoMQ is a cloud-native streaming platform and Apache Kafka distribution that implements a decoupled compute and storage architecture. It functions as an S3-backed message queue, using object storage as the primary log repository to eliminate dependencies on local disks. The platform utilizes a stateless broker architecture to enable dynamic compute scaling and automated partition balancing. This design allows the system to adjust the number of brokers in seconds and distribute network traffic without requiring manual data migration or partition reassignment. The system provides multi-avail

Build data pipelines with SQL and Python, ingest data from different sources, add quality checks, and build end-to-end flows.

ingestr is a command-line tool for copying and syncing data between different database engines and third-party platforms without writing custom code. It functions as an ETL pipeline utility that extracts data from diverse sources and loads it into destinations. The tool features a schema-agnostic data loader that maps source fields to destination columns dynamically, removing the need for predefined static table definitions. It also operates as an incremental data synchronizer, updating destination tables by appending new records or merging changes to maintain current datasets. The system pr

Celery is an asynchronous job processor and distributed task queue designed to offload time-consuming operations to background worker nodes. By utilizing a message-passing architecture, it decouples task producers from consumers, allowing applications to maintain responsiveness while scaling workloads across multiple isolated environments. The system functions as a distributed workload orchestrator that manages the lifecycle of deferred operations through persistent queues. It distinguishes itself by providing a pluggable transport abstraction, which allows the core task logic to remain indep

Python task queue using Redis

Cocoindex is an incremental data processing engine that builds and maintains live indexes for AI agents, with a core focus on codebase indexing and knowledge graph extraction. The engine uses a function-graph execution model where user-defined Python functions are composed into a directed acyclic graph, and it processes data incrementally so only changed source records or code paths are re-computed, avoiding full recomputation at any scale. It supports automatic schema inference from transformation pipeline type annotations and provides full data lineage tracing, tagging every output record wi

.. image:: https://media.charlesleifer.com/blog/photos/huey3-logo.png

Unified Interface for Constructing and Managing Workflows on different workflow engines, such as Argo Workflows, Tekton Pipelines, and Apache Airflow.

CuPy is a CUDA array computing library that implements a NumPy-compatible interface for executing array operations and numerical computing on NVIDIA GPUs. It serves as a GPU-accelerated numerical library and a CUDA-based SciPy implementation, offloading heavy calculations to graphics hardware to increase processing speed for scientific and engineering workloads. The library enables multi-framework tensor exchange, allowing data buffers to be shared between different deep learning frameworks using standardized memory layouts to avoid memory copies. It also supports custom GPU kernel integratio

Dagster is a data orchestration platform designed to manage the entire lifecycle of data assets through declarative modeling and version-controlled code. It functions as a workflow engine that treats data assets as first-class primitives, allowing teams to define, schedule, and monitor complex pipelines while maintaining clear visibility into lineage, dependencies, and data quality. The platform distinguishes itself by using a code-as-configuration framework that enables standard software engineering practices, such as unit testing and local mocking, to be applied directly to data workflows.

Apache Hamilton helps data scientists and engineers define testable, modular, self-documenting dataflows, that encode lineage/tracing and metadata. Runs and scales everywhere python does.

Dask is a parallel computing framework and distributed task scheduler designed to scale Python data science workflows from single machines to large clusters. It functions as a cluster resource manager that orchestrates computational logic by representing tasks and their dependencies as directed acyclic graphs. This architecture allows the system to automate the distribution of workloads across available hardware while managing complex execution requirements. The project distinguishes itself through a lazy evaluation engine that defers data operations until they are explicitly requested, enabl

DBND an open source framework for building and tracking data pipelines. DBND is used for processes ranging from data ingestion, preparation, machine learning model training and production.

Useful extensions to the standard Python datetime features

dbt-core is a command-line framework for transforming data within a warehouse using modular SQL and version control. It functions as a data transformation engine that enables users to define data structures and business logic through declarative configuration files, which the system then compiles into executable code. By managing complex data dependencies through a directed acyclic graph, it ensures that transformation tasks execute in the correct order while maintaining a manifest-driven state to track lineage and execution history. The project distinguishes itself through an adapter-based d

DeepLake is AI data infrastructure consisting of a multimodal data lake, a hybrid search engine, and a serverless vector database. It provides a PostgreSQL-based AI data runtime that combines multimodal storage with streaming pipelines to load and shuffle datasets from cloud storage directly into deep learning training pipelines. The system utilizes lazy indexing to store and slice images, audio, and video without loading entire files into memory. It enables retrieval-augmented generation by persisting high-dimensional embeddings in a serverless vector store and implementing hybrid search tha