Parquet ADO.NET Provider

SQL-based Access to Parquet through ADO.NET for your custom .NET applications and SSAS.

The Parquet ADO.NET Data Provider enables user to easily connect to Parquet data from .NET applications. Rapidly create and deploy powerful .NET applications that integrate with Parquet.

Parquet

Parquet Connectivity Features

SQL access to Apache Parquet data
Connect to live Apache Parquet data, for real-time data access
Full support for data aggregation and complex JOINs in SQL queries
Secure connectivity through modern cryptography, including TLS 1.2, SHA-256, ECC, etc.
Generate table schema automatically based on existing Apache Parquet data or manually for greater control of the content you need
Seamless integration with leading BI, reporting, and ETL tools and with custom applications

Target Service, API

The driver provides SQL access to Apache Parquet files. Supports local files, network paths, and cloud storage locations (S3, Azure Blob, etc.).

Schema, Data Model

Automatically infers schema from Parquet file metadata. Supports complex types including nested structures and arrays.

Key Objects

Parquet files are exposed as tables. Each file or set of files in a directory becomes a queryable table. Supports partitioned datasets.

Operations

Read-only SQL queries on Parquet data. Supports filtering, projections, and aggregations with predicate pushdown for optimal performance.

Authentication

File system or cloud storage authentication. For cloud storage, supports AWS credentials, Azure credentials, or Google Cloud credentials.

See what you can do with Parquet ADO.NET provider

SSAS Cube

Use Parquet from SQL Server Analysis Service (SSAS) multi-dimensional cubes. Keep your analytical data modeling and access to any source including cloud and on-premises.

Custom .NET Application

The Parquet ADO.NET Provider allows developers to build applications that connect to Parquet using familiar SQL and Entity Framework. Integrate Parquet to your mission -critical applications or create easy side-by-side applications.

Low-Code Dev Platforms

You can connect from ADO.NET compliant low-code development tools:

Reporting Tools

You can connect Parquet from .NET-based reporting and analytics tools:

Standard ADO.NET Access to Parquet

The Parquet ADO.NET Provider offers the most natural way to access Parquet data from any .NET application. Simply use Parquet Data Provider objects to connect and access data just as you would access any traditional database. You will be able to use the Parquet Data Provider through Visual Studio Server Explorer, in code through familiar classes, and in data controls like DataGridView, GridView, DataSet, etc.

The CData ADO.NET Provider for Parquet hides the complexity of accessing data and provides additional powerful security features, smart caching, batching, socket management, and more.

Working with DataAdapters, DataSets, DataTables, etc.

The Parquet Data Provider has the same ADO.NET architecture as the native .NET data providers for SQL Server and OLEDB, including: ParquetConnection, ParquetCommand, ParquetDataAdapter, ParquetDataReader, ParquetDataSource, ParquetParameter, etc. Because of this you can now access Parquet data in an easy, familiar way.

For example:

using (ParquetConnection conn = new ParquetConnection("...")) {
	string select = "SELECT * FROM ParquetData";
	ParquetCommand cmd = new ParquetCommand(select, conn);
	ParquetDataAdapter adapter = new ParquetDataAdapter(cmd);
	using (adapter) {
		DataTable table = new DataTable();
		adapter.Fill(table);		
		...
	}
}

ADO.NET Provider Performance

With traditional approaches to remote access, performance bottlenecks can spell disaster for applications. Regardless if an application is created for internal use, a commercial project, web, or mobile application, slow performance can rapidly lead to project failure. Accessing data from any remote source has the potential to create these problems. Common issues include:

Network Connections - Slow network connections and latency issues are common in mobile applications.
Service Delays - Delays due to service interruptions, resulting in server hardware or software updates.
Large Data - Intentional or unintentional requests for large amounts of data.
Disconnects - Complete loss of network connectivity.

The CData ADO.NET Provider for Parquet solves these issues by supporting powerful smart caching technology that can greatly improve the performance and dramatically reduce application bottlenecks.

Smart Caching

Smart caching is a configurable option that works by storing queried data into a local database. Enabling smart caching creates a persistent local cache database that contains a replica of data retrieved from the remote source. The cache database is small, lightweight, blazing-fast, and it can be shared by multiple connections as persistent storage.

Caching with our ADO.NET Providers is highly configurable, including options for:

Auto Cache - Maintain an automatic local cache of data on all requests. The provider will automatically load data into the cache database each time you execute a SELECT query. Each row returned by the query will be inserted or updated as necessary into the corresponding table in the cache database.
Explicit Cache - Cache only on demand. Developers decide exactly what data gets stored in the cache and when it is updated. Explicit caching provides full control over the cache contents by using explicit execution of CACHE statements.
No Cache - All requests access only live data and no local cache file is created.

This powerful caching functionality increases application performance and allows applications to disconnect and continue limited functioning without writing code for additional local storage and/or data serialization/deserialization.

More information about ADO.NET Provider caching and best caching practices is available in the included help files.

Visual Studio Integration & Server Explorer

Working with the new Parquet ADO.NET Provider is easy. As a fully-managed .NET Data Provider, the Parquet Data Provider integrates seamlessly with the Visual Studio development environment as well as any .NET application.

As an ADO.NET Data Provider, Parquet ADO.NET Provider can be used to access and explore Apache Parquet data directly from the Visual Studio Server Explorer.

It's easy. As a standard ADO.NET adapter, developers can connect the Server Explorer to Parquet ADO.NET Provider just like connecting to any standard database.

Add a new Data Connection from the Server Explorer and select the Apache Parquet Data Source
Configure the basic connection properties to access your Apache Parquet account data.

Explore all of the data available! Parquet ADO.NET Provider makes it easy to access live Apache Parquet data from Visual Studio.

After configuring the connection, explore the feeds, views, and services provided by the Apache Parquet Data Source.
These constructs return live Apache Parquet data that developers can work with directly from within Visual Studio!

Developer Integration: Databind to Parquet

Connecting Web, Desktop, and Mobile .NET applications with Apache Parquet is just like working with SQL Server. It is even possible to integrate Parquet ADO.NET Provider into applications without writing code.

Developers are free to access the Parquet ADO.NET Provider in whatever way they like best. Either visually through the Visual Studio Winforms or Webforms designers, or directly through code.

Developers can connect the Apache Parquet Data Source directly to form components by configuring the object's smart tags.
Add a new Data Connection from the Server Explorer and select the Apache Parquet Data Source. Then, select the feed, view, or services you would like to connect the object to.

Done! It's just like connecting to SQL Server.

Once the object is bound to the data source, applications can easily interact with Apache Parquet data with full read/write (CRUD) support.