index.html

---
layout: default
title: DuckLake is an integrated data lake and catalog format
subtitle: An integrated data lake and catalog format
description: DuckLake delivers advanced data lake features without traditional lakehouse complexity by using Parquet files and your SQL database. It's an open, standalone format from the DuckDB team.
body_class: landing nowrap
---

<section class="welcome">
	<div class="wrap">
		<div class="ducklake">
			<h1>DuckLake is an integrated data&nbsp;lake and catalog format</h1>
			<p>DuckLake delivers advanced data&nbsp;lake features without traditional lakehouse complexity by using Parquet files and your SQL database. It's an open, standalone format from the DuckDB team.</p>
			<div class="cta">
				<a class="button black" href="#quickinstall">Get started</a>
				<a class="button transparent" href="{% link docs/stable/index.md %}">Documentation</a>
			</div>
		</div>
	</div>
</section>

<div class="wave"></div>

<section class="architecture">
	<div class="wrap">
		<div class="section-text">
			<h2>DuckLake’s architecture</h2>
			<p>With DuckLake, all you need to run your own data warehouse is a catalog database and storage for Parquet files.</p>
		</div>
		<div class="illustration">
				<div class="window architecture-tabs">
					<div class="topbar">
						<ul>
							<div class="select-highlight"></div>
							<li data-tab="arch-tab2" data-iconclass="postgre" data-multi="true"  class="active">PostgreSQL</li>
							<li data-tab="arch-tab3" data-iconclass="sqlite" data-multi="true">SQLite</li>
							<li data-tab="arch-tab4" data-iconclass="mysql" data-multi="true">MySQL</li>
							<li data-tab="arch-tab1" data-iconclass="duckdb" data-multi="false">DuckDB</li>
						</ul>
						<div class="hoverlabel">&#8592; Choose catalog database</div>
					</div>
					
				
					<div class="diagram">
						<div class="client">
							<h4>Client</h4>
							
							<div class="browser-wrap contains-multi">
								<div class="browser">
									<img src="{{ site.baseurl }}/images/deployment_diagram/window.svg" alt="Database Icon">
									<div class="broser-content">
										<div class="ducklake">
											<img src="{{ site.baseurl }}/images/deployment_diagram/ducklake_comb.svg" class="ducklake_icon">
											<!--<div class="sub">DuckLake client</div>-->
										</div>
									</div>
								</div>
								<div class="browser multi">
									<img src="{{ site.baseurl }}/images/deployment_diagram/window.svg" alt="Database Icon">
									
								</div>
								
							</div>
							
							<div class="hover-content">
								<div class="mysql postgre sqlite">
									<h4>Clients</h4>
									<p>Users can run multiple DuckLake clients and connect concurrently to the catalog database – PostgreSQL, MySQL or SQLite – to work over the same DuckLake dataset.</p>
								</div>
								<div class="duckdb">
									<h4>Client</h4>
									<p>DuckLake also works with DuckDB as the catalog database. In this case, you are limited to a single client but still get to enjoy the data lake features such as time travel.</p>
								</div>
							</div>
							
						</div>
						<div class="database-storage">
							<h4>DuckLake</h4>
							<div class="database">
								<h4>Catalog: <span class="choosen-catalog"></span></h4>
								
								<div class="catalog">
									<img src="{{ site.baseurl }}/images/deployment_diagram/database.svg" alt="Database Icon">
									<div class="icons">
										<img src="{{ site.baseurl }}/images/deployment_diagram/cat_icon-duckdb.svg" class="duckdb">
										<img src="{{ site.baseurl }}/images/deployment_diagram/cat_icon-postgre.svg" class="postgre">
										<img src="{{ site.baseurl }}/images/deployment_diagram/cat_icon-sqlite.svg" class="sqlite">
										<img src="{{ site.baseurl }}/images/deployment_diagram/cat_icon-mysql.svg" class="mysql">
									</div>
								</div>
								
								<div class="hover-content">
									<div class="mysql postgre duckdb sqlite">
										<h4>Catalog</h4>
										<p>DuckLake can use any SQL database as its catalog, provided that it supports ACID transactions and primary key constraints.</p>
									</div>
								</div>
								
								
							</div>
							<div class="storage">
								<h4>Storage: <span class="choosen-storage">Parquet</span></h4>
								
								<div class="folder">
									<img src="{{ site.baseurl }}/images/deployment_diagram/parquet_folder.svg" alt="Parquet Folder">
								</div>
								
								<div class="hover-content">
									<div class="mysql postgre sqlite duckdb">
										<h4>Storage</h4>
										<p>DuckLake can store its data as Parquet files on a local disk, or in an object storage such as AWS S3 or Google Cloud Storage.</p>
									</div>
								</div>
							</div>
							<div class="connector">
								<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 54 9" fill="none">
								  <path fill="#1A1A1A" d="M50 .5a4 4 0 1 1 0 8 4 4 0 0 1 0-8ZM4 .5a4 4 0 1 1 0 8 4 4 0 0 1 0-8Zm46 4v.75H4v-1.5h46v.75Z" />
								</svg>
							</div>
						</div>
					</div>
					
					
				</div>
		</div>
	</div>
</section>


<section class="features">
	<div class="wrap">
		<h2>DuckLake’s key features</h2>
		<div class="cards vertical icons">
			<div class="card">
				<div class="icon">
					<img src="{{ site.baseurl }}/images/icons/waves.svg" alt="wave icon">
				</div>
				<div class="content">
					<h3>Data lake operations</h3>
					<p>DuckLake supports snapshots, time travel queries, schema evolution and partitioning.</p>
				</div>
			</div>
			<div class="card">
				<div class="icon">
					<img src="{{ site.baseurl }}/images/icons/documents.svg" alt="wave icon">
				</div>
				<div class="content">
					<h3>Lightweight snapshots</h3>
					<p>You can have as many snapshots as you want without frequent compacting steps!</p>
				</div>
			</div>
			<div class="card">
				<div class="icon">
					<img src="{{ site.baseurl }}/images/icons/pipette.svg" alt="wave icon">
				</div>
				<div class="content">
					<h3>ACID transactions</h3>
					<p>DuckLake allows concurrent access with ACID transactional guarantees over multi-table operations.</p>
				</div>
			</div>
			<div class="card">
				<div class="icon">
					<img src="{{ site.baseurl }}/images/icons/clock.svg" alt="wave icon">
				</div>
				<div class="content">
					<h3>Performance-oriented</h3>
					<p>DuckLake uses statistics for filter pushdown, enabling fast queries even on large datasets.</p>
				</div>
			</div>
		</div>
	</div>
</section>

<section class="podcast">
	<div class="wrap">
		<div class="section-text">
			<h2>In conversation: DuckDB founders on DuckLake</h2>
			<p>Listen to Hannes Mühleisen and Mark Raasveldt walk through the history of data lakes and introduce DuckLake, a new lakehouse format.</p>
		</div>
		<div class="section-image">
			<div class="video-thumbnail" data-video-id="zeonmOO9jm4">
				<img src="{{ site.baseurl }}/images/thumb_introducting-ducklake.png" alt="Thumbnail: Introducing DuckLake">
				<div class="play-button">
					<svg xmlns="http://www.w3.org/2000/svg" width="102" height="102" fill="none"><path fill="#000" fill-opacity=".2" d="M51 0C22.877 0 0 22.877 0 51s22.877 51 51 51 51-22.877 51-51S79.123 0 51 0Z"/><path fill="#99D8FF" d="M71.442 49.21 43.319 29.975c-.874-.645-3.206-.645-3.393 1.79v38.47a2.158 2.158 0 0 0 3.393 1.79L71.442 52.79c.728-.437 1.728-2.227 0-3.58Z"/></svg>
				</div>
			</div>
			<div class="video-container" style="display: none;">
			  <iframe data-src="https://www.youtube.com/embed/zeonmOO9jm4" title="YouTube video player" width="560" height="315" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
			</div>
		</div>
	</div>
</section>

{% include quick_installation.html %}

<section class="faq">
	<div class="wrap">
		<div class="section-text">
			<h2>Frequently asked questions</h2>
			<p>Answers to common questions to help you understand and make the most of DuckLake.</p>
		</div>	
	</div>
	
	<div class="frequentlyaskedquestions">
		<div class="qa-wrap">
			<h3>
				Why should I use DuckLake?
			</h3>
			<div class="answer">
				<p>DuckLake provides a lightweight one-stop solution if you need a data lake and catalog.

				<p>You can use DuckLake for a “multiplayer DuckDB” setup with multiple DuckDB instances reading and writing the same dataset –
				a concurrency model <a href="https://duckdb.org/docs/stable/connect/concurrency">not supported by vanilla DuckDB</a>.</p>

				<p>If you only use DuckDB for both your DuckLake entry point and your catalog database, you can still benefit from DuckLake:
				you can run time travel queries,
				exploit data partitioning,
				and can store your data in multiple files instead of using a single (potentially very large) database file.</p>
			</div>
		</div>

		<div class="qa-wrap">
			<h3>
				What is “DuckLake”?
			</h3>
			<div class="answer">
				“DuckLake” can refer to a number of things:

				<ol>
					<li>The <i>DuckLake format</i> that uses a catalog database and a Parquet storage to store data.</li>
					<li>A <i>DuckLake instance</i> storing a dataset with the DuckLake lakehouse format.</li>
					<li>The <a href="{% link docs/stable/duckdb/introduction.md %}"><code>ducklake</code> <i>DuckDB extension</i></a>, which supports reading/writing datasets using the DuckLake format.</li>
				</ol>
			</div>
		</div>

		<div class="qa-wrap">
			<h3>
				What is the license of DuckLake?
			</h3>
			<div class="answer">
				The DuckLake specification and the <code>ducklake</code> DuckDB extension are released under the MIT license.
			</div>
		</div>

	</div>
	
	<div class="cta">
		<a class="button black" href="{% link faq.md %}">More FAQs</a>
	</div>
	
</section>