OPeNDAP

OPeNDAP, short for Open-source Project for a Network Data Access Protocol, is an open scientific-data access project associated with OPeNDAP, Inc. It develops and promotes the Data Access Protocol (DAP), a web-oriented standard and software architecture for accessing structured datasets across networks. Rather than requiring users to download complete files before analysis, DAP allows a client to discover a dataset's metadata and request selected variables, dimensions, records, or other subsets from a remote server. This approach is particularly useful for large scientific collections that are expensive or impractical to transfer in full. The project is strongly associated with Earth and environmental science, where datasets commonly contain multidimensional arrays, time series, geospatial grids, satellite observations, model outputs, and in-situ measurements. DAP is layered on HTTP, allowing data services to operate through familiar web infrastructure. A client identifies a dataset with a URL, requests structural metadata, and then submits a constraint expression describing the information it needs. Servers can return data and metadata in several representations, including XML, JSON, HTML, ASCII, and binary formats designed for efficient scientific exchange. DAP2 remains widely deployed, while DAP4 is the current specification described by the project. The protocol's data model supports structures such as multidimensional arrays and nested sequences, making it applicable to both regular gridded data and more record-oriented scientific information. Subsetting, aggregation, and streaming can reduce network traffic and improve access to large collections, especially when users need only a small portion of a remote resource. OPeNDAP maintains open-source reference software for both server and client use. Its Hyrax server framework, built to operate with Apache, can expose data from sources such as NetCDF and HDF files, databases, and other formats through adapters. The project has also provided libraries in languages including C++ and Java. Other widely used scientific software, including NetCDF libraries and Python-based implementations such as PyDAP, can provide DAP access or interoperate with DAP services. Visualization and analysis applications such as MATLAB, IDL, Panoply, GrADS, Integrated Data Viewer, Ferret, and ncBrowse have been used as examples of clients adapted for remote DAP data. OPeNDAP is part of a broader interoperable ecosystem. Systems such as THREDDS and ERDDAP use related or compatible approaches and can exchange data with OPeNDAP servers. Government and research agencies, including NASA and NOAA, have used DAP-based services to distribute Earth-science information, including satellite and oceanographic datasets. The brand therefore represents both a protocol and a family of open software and service practices for making specialized scientific data discoverable, selectable, and usable over the web. Public information does not establish a conventional consumer market, a listed corporate structure, or a publicly documented founder and headquarters for the organization.

OPeNDAP emerged around the need to make scientific datasets available through network services without forcing every user to retrieve complete files. Conventional file transfer is poorly suited to large observational and simulation datasets: a researcher may need only a few variables, geographic cells, time steps, or records, while the underlying file may contain much more information. The Data Access Protocol addressed this problem by defining a client-server model in which a client can inspect a remote dataset, formulate a constrained request, and receive only the relevant content. The project developed a web-based architecture using HTTP as its transport layer. This choice helped DAP services fit into existing internet and institutional infrastructure while remaining useful to scientific applications rather than only ordinary web browsers. The protocol's data model was designed for scientific structures, including arrays and sequences, and its constraint expressions enabled selective retrieval. Servers could expose data stored in different physical formats through adapters, separating the way data were stored from the way users accessed them. OPeNDAP's role expanded beyond specification work into reference implementations and interoperability support. The organization developed server-side and client-side open-source components, including libraries in C++ and Java. Hyrax became the project's principal server framework, operating with Apache and supporting adapters for sources such as HDF and NetCDF files, databases, and other data formats. These components made it possible for institutions to publish existing scientific collections as network-accessible services rather than redesigning their archives around a single file format. The protocol became especially important in Earth science and related fields. NASA, NOAA, and other government and research organizations use DAP-based services or compatible systems to provide access to high-volume information such as satellite observations, ocean data, and model products. Scientific analysis and visualization tools adapted to DAP could work with remote collections through embedded libraries. The NetCDF software ecosystem also became an important route to DAP access, while community projects supplied implementations in additional programming languages. DAP2 became broadly deployed and remains in use. DAP4 represents the later specification, extending the protocol's capabilities while preserving the project's general goal of structured remote access. In parallel, other data-service architectures, notably THREDDS and ERDDAP, developed complementary or interoperable approaches. Their coexistence helped form a wider ecosystem in which a dataset could be published by one service and consumed by multiple analysis, visualization, and cataloging tools. OPeNDAP is therefore best understood as a technical project and open scientific-infrastructure brand rather than a conventional commercial product brand. Its significance lies in the combination of a protocol, server and client implementations, and a community of interoperable tools. Public reference material does not provide enough reliable information to establish a precise founding year, named founder, headquarters, executive roster, or a sequence of commercial corporate milestones.

An open, discipline-neutral web architecture for efficient remote access to structured scientific data, with particular strength in Earth and environmental science.

• Scientific data-access protocol

  Data Access Protocol (DAP)

  DAP is OPeNDAP's core protocol and architectural model for accessing structured datasets over the web. A client can request metadata, construct a constraint expression, and retrieve selected portions of a remote resource instead of transferring the entire dataset. The model supports multidimensional arrays, sequences, metadata, streaming, and multiple response representations. DAP2 is widely deployed, while DAP4 is the current specification identified with the project.

• DAP server framework

  Hyrax

  Hyrax is OPeNDAP's open-source server software for publishing data through DAP services. It is designed to run on top of Apache and uses adapters to expose information from sources including NetCDF and HDF files, databases, and other data formats. By supporting subsetting and aggregation, Hyrax can present large scientific collections as queryable network resources and can deliver selected data without requiring users to download complete source files.

• Open-source software libraries

  DAP client and server libraries

  OPeNDAP provides reference software and libraries that help applications implement DAP clients or servers. The project has maintained implementations in languages including C++ and Java, while related community and scientific-computing projects provide support in additional languages. These libraries allow analysis tools, visualization programs, and web applications to discover remote datasets, formulate constrained requests, and decode returned scientific content.

1. 1OPeNDAPen.wikipedia.org
2. 2OPeNDAP official websiteopendap.org