4.3k GitHub stars and counting — facebookincubator/velox is a C++ project TopGit is tracking across repositories on the platform. A composable and fully extensible C++ execution engine library for data management systems.
Snapshot summary built from the project's own GitHub metadata — there's no written TopGit review yet. The page will update automatically when a full review is published.
WHY NO REVIEW YET
TopGit writes full reviews for the most-starred, most-requested repositories. This page is a snapshot until then — see the READ ME tab for the original README in full.
Velox is a composable execution engine distributed as an open source C++
library. It provides reusable, extensible, and high-performance data processing
components that can be (re-)used to build data management systems focused on
different analytical workloads, including batch, interactive, stream
processing, and AI/ML. Velox was created by Meta and it is currently developed
in partnership with IBM/Ahana, Intel, Voltron Data, Microsoft, ByteDance and
many other companies.
In common usage scenarios, Velox takes a fully optimized query plan as input
and performs the described computation. Considering Velox does not provide a
SQL parser, a dataframe layer, or a query optimizer, it is usually not meant
to be used directly by end-users; rather, it is mostly used by developers
integrating and optimizing their compute engines.
Velox provides the following high-level components:
Type: a generic typing system that supports scalar, complex, and nested
types, such as structs, maps, arrays, etc.
Vector: an Arrow-compatible columnar memory layout
module,
providing encodings such as Flat, Dictionary, Constant, and Sequence/RLE, in
addition to a lazy materialization pattern and support for out-of-order
writes.
Expression Eval: a fully vectorized expression evaluation
engine
that allows expressions to be efficiently executed on top of Vector/Arrow
encoded data.
Functions: sets of vectorized scalar, aggregates, and window functions
implementations following the Presto and Spark semantic.
Operators: implementation of relational operators such as scans, writes,
projections, filtering, grouping, ordering, shuffle/exchange, hash, merge,
and nested loop joins,
unnest, and more.
I/O: a connector interface for extensible data sources and sinks,
supporting different file formats (ORC/DWRF, Parquet, Nimble), and storage
adapters (S3, HDFS, GCS, ABFS, local files) to be used.
Network Serializers: an interface where different wire protocols can be
implemented, used for network communication, supporting
PrestoPage
and Spark's UnsafeRow.
Resource Management: a collection of primitives for handling
computational resources, such as memory
arenas and
buffer management, tasks, drivers, and thread pools for CPU and thread
execution, spilling, and caching.
Velox is extensible and allows developers to define their own engine-specific
specializations, including:
Custom types
Simple and vectorized functions
Aggregate functions
Window functions
Operators
File formats
Storage adapters
Network serializers
Examples
Examples of extensibility and integration with different component APIs can be
found here
Documentation
Developer guides detailing many aspects of the library, in addition to the list
of available functions can be found here.
Recent blog posts (all posts):
Iceberg V3 in Velox (2026-10-30)
Hybrid FlatMap: Group-Level Feature Projection in Nimble (2026-10-09)
Parallel Encoding and Decoding in Nimble: Faster Flushes and Reads (2026-10-05)
Community
Velox is an open source project supported by a community of individual
contributors and organizations. The project's technical governance mechanics is
described in this
document..
Project maintainers are listed
here.
The main communication channel with the Velox OSS community is through the the
Velox-OSS Slack workspace, github Issues, and
Discussions.
For access to the Velox Slack workspace, please add a comment to this
Discussion
Contributing
Check our contributing guide to learn about how to
contribute to the project.
License
Velox is licensed under the Apache 2.0 License. A copy of the license
can be found here.
Getting Started
Get the Velox Source
git clone https://github.com/facebookincubator/velox.git
cd velox
Once Velox is checked out, the first step is to install the dependencies.
Details on the dependencies and how Velox manages some of them for you
can be found here.
Velox also provides the following scripts to help developers setup and install Velox
dependencies for a given platform.
Supported OS and compiler matrix
The minimum versions of supported compilers:
OS
Compiler
Version
Linux
gcc
11
Linux
clang
15
macOS
clang
15
The recommended OS versions and compilers:
OS
Compiler
Version
CentOS 9/RHEL 9
gcc
12
Ubuntu 22.04
gcc
11
macOS
clang
16
Alternative combinations:
OS
Compiler
Version
CentOS 9/RHEL 9
gcc
11
Ubuntu 20.04
gcc
11
Ubuntu 24.04
clang
15
CentOS 9/RHEL 9
clang
system
Setting up dependencies
The following setup scripts use the DEPENDENCY_DIR environment variable to set the
location to download and build packages. This defaults to deps-download in the current
working directory.
Use INSTALL_PREFIX to set the install directory of the packages. This defaults to
deps-install in the current working directory on macOS and to the default install
location (eg. /usr/local) on linux.
Using the default install location /usr/local on macOS is discouraged since this
location is used by certain Homebrew versions.
Manually add the INSTALL_PREFIX value in the IDE or bash environment,
say export INSTALL_PREFIX=/Users/$USER/velox/deps-install to ~/.zshrc so that
subsequent Velox builds can use the installed packages.
You can reuse DEPENDENCY_INSTALL and INSTALL_PREFIX for Velox clients such as Prestissimo
by specifying a common shared directory.`
The build parallelism for dependencies can be controlled by the BUILD_THREADS environment
variable and overrides the default number of parallel processes used for compiling and linking.
The default value is the number of cores on your machine.
This is useful if your machine has lots of cores but no matching memory to process all
compile and link processes in parallel resulting in OOM kills by the kernel.
Setting up on macOS
On a macOS machine (either Intel or Apple silicon) you can setup and then build like so:
$ ./scripts/setup-macos.sh
$ make
With macOS 14.4 and XCode 15.3 where m4 is missing, you can either
The supported architectures are x86_64 (avx, sse), and AArch64 (apple-m1+crc, neoverse-n1).
You can build like so:
$ ./scripts/setup-ubuntu.sh
$ make
Setting up on Centos 9 Stream with adapters
Velox adapters include file-systems such as AWS S3, Google Cloud Storage,
and Azure Blob File System. These adapters require installation of additional
libraries. Once you have checked out Velox, you can setup and build like so:
$ ./scripts/setup-centos9.sh
$ ./scripts/setup-centos9.sh install_adapters
$ make
Note that the install_adapters command is available for the supported MacOS and
Ubuntu (20.04 or later) scripts. Individual adapters can be installed by specifying
the individual install command, e.g. setup-centos9.sh install_aws.
Using Clang on Linux
Clang can be additionally installed during the setup step for Ubuntu 22.04/24.04
and CentOS 9 by setting the USE_CLANG environment variable prior to running the platform specific setup script.
$ export USE_CLANG=true
This will install and use Clang to build the dependencies instead of using the default GCC compiler:
Clang 15 on Ubuntu, and the system Clang with its OpenMP runtime on CentOS 9.
Once completed, and before running any make command, set the compiler to be used. On Ubuntu:
$ export CC=/usr/bin/clang-15
$ export CXX=/usr/bin/clang++-15
$ make
On CentOS 9:
$ export CC=/usr/bin/clang
$ export CXX=/usr/bin/clang++
$ make
Building Velox
Run make in the root directory to compile the sources. For development, use
make debug to build a non-optimized debug version, or make release to build
an optimized version. Use make unittest to build and run tests.
Four test suites use grouped binaries on Linux CI to reduce link times
(velox/exec/tests, velox/functions/prestosql/aggregates/tests,
velox/common/caching/tests, velox/serializers/tests). All other suites use
individual binaries on all platforms. On macOS, grouping is off by default. To
disable grouping on Linux, pass -DVELOX_ENABLE_GROUPED_TESTS=OFF via
EXTRA_CMAKE_FLAGS.
Note that,
Velox requires a compiler at the minimum GCC 11.0 or Clang 15.0.
On x86 CPUs, Velox requires at least SSE4.2. By default, Velox builds for
CPU_TARGET=avx on AVX-capable hosts, which requires AVX, AVX2, FMA, F16C,
LZCNT, and BMI2. To build for the minimum x86 baseline, set
CPU_TARGET=sse.
On Arm CPUs, Velox supports Apple Silicon with CPU_TARGET=arm64 and Linux
AArch64 with CPU_TARGET=aarch64.
Build metrics for Velox are published at https://facebookincubator.github.io/velox/bm-report/
Building Velox with docker-compose
If you don't want to install the system dependencies required to build Velox,
you can also build and run tests for Velox on a docker container
using docker-compose.
Use the following commands:
$ docker-compose build ubuntu-cpp
$ docker-compose run --rm ubuntu-cpp
If you want to increase or decrease the number of threads used when building Velox
you can override the NUM_THREADS environment variable by doing:
$ docker-compose run -e NUM_THREADS=<NUM_THREADS_TO_USE> --rm ubuntu-cpp
How active is development on facebookincubator/velox?
The most recent commit recorded on facebookincubator/velox was today, based on the GitHub push timestamp. The repository has 1.7k forks — one of the better signals of community interest.
How many stars does facebookincubator/velox have?
facebookincubator/velox has 4.3k GitHub stars — refresh the page for the live number, or check github.com/facebookincubator/velox. TopGit mirrors GitHub's count but does not claim minute-by-minute accuracy.
What is facebookincubator/velox?
facebookincubator/velox (facebookincubator/velox) is a C++ project on GitHub. From the project's own README: A composable and fully extensible C++ execution engine library for data management systems.
What language is facebookincubator/velox written in?
facebookincubator/velox is written primarily in C++. GitHub's language field is based on the largest share of bytes in the default branch.
What topics is facebookincubator/velox associated with?
GitHub's repository topics for facebookincubator/velox: "data-management", "query-processing". TopGit's editorial category is open-source.
Read full README in the tab above.
Curious whether velox is right for you?
Let ChatGPT, Claude, or Perplexity look into it — click below and see what AI actually says about velox.