Microsoft SEAL is an easy-to-use open-source (MIT licensed) homomorphic encryption library developed by the Cryptography and Privacy Research group at Microsoft. Microsoft SEAL is written in modern standard C++ and has no external dependencies, making it easy to compile and run in many different environments. For more information about the Microsoft SEAL project, see sealcrypto.org.
This document pertains to Microsoft SEAL version 3.4. Users of previous versions of the library should look at the list of changes.
- Introduction
- Installing Microsoft SEAL
- Enabling Optional Dependencies
- Building Microsoft SEAL for .NET
- Getting Started
- Contributing
- Citing Microsoft SEAL
Most encryption schemes consist of three functionalities: key generation, encryption, and decryption. Symmetric-key encryption schemes use the same secret key for both encryption and decryption; public-key encryption schemes use separately a public key for encryption and a secret key for decryption. Therefore, public-key encryption schemes allow anyone who knows the public key to encrypt data, but only those who know the secret key can decrypt and read the data. Symmetric-key encryption can be used for efficiently encrypting very large amounts of data, and enables secure outsourced cloud storage. Public-key encryption is a fundamental concept that enables secure online communication today, but is typically much less efficient than symmetric-key encryption.
While traditional symmetric- and public-key encryption can be used for secure storage and communication, any outsourced computation will necessarily require such encryption layers to be removed before computation can take place. Therefore, cloud services providing outsourced computation capabilities must have access to the secret keys, and implement access policies to prevent unauthorized employees from getting access to these keys.
Homomorphic encryption refers to encryption schemes that allow the cloud to compute directly on the encrypted data, without requiring the data to be decrypted first. The results of such encrypted computations remain encrypted, and can be only decrypted with the secret key (by the data owner). Multiple homomorphic encryption schemes with different capabilities and trade-offs have been invented over the past decade; most of these are public-key encryption schemes, although the public-key functionality may not always be needed.
Homomorphic encryption is not a generic technology: only some computations on encrypted data are possible. It also comes with a substantial performance overhead, so computations that are already very costly to perform on unencrypted data are likely to be infeasible on encrypted data. Moreover, data encrypted with homomorphic encryption is many times larger than unencrypted data, so it may not make sense to encrypt, e.g., entire large databases, with this technology. Instead, meaningful use-cases are in scenarios where strict privacy requirements prohibit unencrypted cloud computation altogether, but the computations themselves are fairly lightweight.
Typically, homomorphic encryption schemes have a single secret key which is held by the data owner. For scenarios where multiple different private data owners wish to engage in collaborative computation, homomorphic encryption is probably not a reasonable solution.
Homomorphic encryption cannot be used to enable data scientist to circumvent GDPR. For example, there is no way for a cloud service to use homomorphic encryption to draw insights from encrypted customer data. Instead, results of encrypted computations remain encrypted and can only be decrypted by the owner of the data, e.g., a cloud service customer.
Microsoft SEAL is a homomorphic encryption library that allows additions and multiplications to be performed on encrypted integers or real numbers. Other operations, such as encrypted comparison, sorting, or regular expressions, are in most cases not feasible to evaluate on encrypted data using this technology. Therefore, only specific privacy-critical cloud computation parts of programs should be implemented with Microsoft SEAL.
It is not always easy or straightfoward to translate an unencrypted computation into a computation on encrypted data, for example, it is not possible to branch on encrypted data. Microsoft SEAL itself has a steep learning curve and requires the user to understand many homomorphic encryption specific concepts, even though in the end the API is not too complicated. Even if a user is able to program and run a specific computation using Microsoft SEAL, the difference between efficient and inefficient implementations can be several orders of magnitude, and it can be hard for new users to know how to improve the performance of their computation.
Microsoft SEAL comes with two different homomorphic encryption schemes with very different properties. The BFV scheme allows modular arithmetic to be performed on encrypted integers. The CKKS scheme allows additions and multiplications on encrypted real or complex numbers, but yields only approximate results. In applications such as summing up encrypted real numbers, evaluating machine learning models on encrypted data, or computing distances of encrypted locations CKKS is going to be by far the best choice. For applications where exact values are necessary, the BFV scheme is the only choice.
Microsoft SEAL comes with a Microsoft Visual Studio 2019 solution file SEAL.sln
that can be used to conveniently build the library, examples, and unit tests. Visual
Studio 2017 version 15.3 or newer is required to build Microsoft SEAL.
The Visual Studio solution SEAL.sln is configured to build Microsoft SEAL both
for Win32 and x64 platforms. Please choose the right platform before building
Microsoft SEAL. The SEALNetNative project or the .NET wrapper library SEALNet
can only be built for x64.
You can easily switch from Visual Studio build configuration menu whether Microsoft
SEAL should be built in Debug mode (no optimizations) or in Release mode. Please
note that Debug mode should not be used except for debugging Microsoft SEAL itself,
as the performance will be orders of magnitude worse than in Release mode.
Build the SEAL project native\src\SEAL.vcxproj from SEAL.sln. This results in
the static library seal.lib to be created in native\lib\$(Platform)\$(Configuration).
When linking with applications, you need to add native\src\ (full path) as an
include directory for Microsoft SEAL header files.
Build the SEALExamples project native\examples\SEALExamples.vcxproj from SEAL.sln.
This results in an executable sealexamples.exe to be created in
native\bin\$(Platform)\$(Configuration).
The unit tests require the Google Test framework to be installed. The appropriate
NuGet package is already listed in native\tests\packages.config, so once you
attempt to build the SEALTest project native\tests\SEALTest.vcxproj from SEAL.sln
Visual Studio will automatically download and install it for you.
Microsoft SEAL is very easy to configure and build in Linux and macOS using CMake (>= 3.12). A modern version of GNU G++ (>= 6.0) or Clang++ (>= 5.0) is needed. In macOS the Xcode toolchain (>= 9.3) will work.
In macOS you will need CMake with command line tools. For this, you can either
- install the cmake package with Homebrew, or
- download CMake directly from cmake.org/download and enable command line tools.
Below we give instructions for how to configure, build, and install Microsoft SEAL either system-wide (global install), or for a single user (local install). A system-wide install requires elevated (root) privileges.
You can easily switch from CMake configuration options whether Microsoft SEAL should be
built in Debug mode (no optimizations) or in Release mode. Please note that Debug
mode should not be used except for debugging Microsoft SEAL itself, as the performance
will be orders of magnitude worse than in Release mode.
We assume that Microsoft SEAL has been cloned into a directory called SEAL and all
commands presented below are assumed to be executed in the directory SEAL.
You can build Microsoft SEAL for your machine by executing the following commands:
cd native/src
cmake .
make
cd ../..
After building Microsoft SEAL, you can build the examples as follows:
cd native/examples
cmake .
make
cd ../..
The sealexamples executable can now be found in native/bin/.
To build the unit tests you will need the GoogleTest framework, which is included in Microsoft SEAL as a git submodule. To download the GoogleTest source files, do:
git submodule update --init
This needs to be executed only once, and can be skipped if Microsoft SEAL was cloned with
git --recurse-submodules. To build the tests, do:
cd native/tests
cmake .
make
cd ../..
The sealtest executable can now be found in native/bin/. All unit tests should pass
successfully.
If you have root access to the system you can install Microsoft SEAL system-wide as follows:
cd native/src
cmake .
make
sudo make install
cd ../..
To instead install Microsoft SEAL locally, e.g., to ~/mylibs/, do the following:
cd native/src
cmake . -DCMAKE_INSTALL_PREFIX=~/mylibs
make
make install
cd ../..
It is very easy to link your own applications and libraries with Microsoft SEAL if
you use CMake. Simply add the following to your CMakeLists.txt:
find_package(SEAL 3.4.1 EXACT REQUIRED)
target_link_libraries(<your target> SEAL::seal)
If Microsoft SEAL was installed globally, the above find_package command will likely
find the library automatically. To link with a locally installed Microsoft SEAL, e.g.,
installed in ~/mylibs as described above, you may need to tell CMake where to look for
Microsoft SEAL when you configure your application by running:
cd <directory containing your CMakeLists.txt>
cmake . -DCMAKE_PREFIX_PATH=~/mylibs
Microsoft SEAL has no required dependencies, but certain optional features can be enabled if it is compiled with support for specific third-party libraries.
Microsoft GSL (Guidelines Support Library) is a header-only library that implements
two convenient (templated) data types: gsl::span and gsl::multi_span. These
are view types that provide safe (bounds-checked) array access to memory. For
example, if Microsoft GSL is available, Microsoft SEAL can allow BatchEncoder
and CKKSEncoder to encode from and decode to a gsl::span instead of std::vector,
which can have significant benefit in performance. Additionally, BatchEncoder allows
access to the slot data alternatively through a two-dimensional gsl::multi_span,
reflecting the batching slot structure. Also the Ciphertext class allows the
ciphertext data to be accessed hierarchically through a gsl::multi_span.
To build Microsoft SEAL with support for Microsoft GSL, clone first the Microsoft GSL
library from GitHub.com/Microsoft/GSL to some
convenient directory, e.g., C:\MyLibs\GSL in this example.
Next, you will need to signal Microsoft SEAL to enable Microsoft GSL support by
creating a new Windows environment variable MSGSL_ROOT, and setting its value to
C:\MyLibs\GSL\include. Restart Visual Studio at this point if you had it open,
otherwise it will not have captured the newly created environment variable.
Rebuilding Microsoft SEAL should now automatically detect that Microsoft GSL is
available, and enable both gsl::span and gsl::multi_span support. To disable
Microsoft GSL support, delete the MSGSL_ROOT environment variable, restart Visual
Studio, and rebuild Microsoft SEAL.
If Microsoft SEAL is built with Microsoft GSL support, any programs or libraries
consuming Microsoft SEAL will need access to the Microsoft GSL header files, so you
need to add $(MSGSL_ROOT) to Additional Include Directories under the C/C++ tab
in your Visual Studio project properties. Note that in the Microsoft SEAL projects
this has already been set for you, so all projects in SEAL.sln should work without
change.
On some Linux distributions Microsoft GSL can be conveniently obtained through
a package manager, e.g., on Ubuntu it suffices to install the package libmsgsl-dev.
Alternatively, you can simply clone it from
GitHub.com/Microsoft/GSL. When installed with
a package manager, CMake will likely detect the Microsoft GSL location automatically.
Otherwise, if Microsoft GSL is cloned to ~/mylibs/GSL, you need to provide CMake
with this location when building Microsoft SEAL as follows:
cd native/src
cmake . -DMSGSL_ROOT=~/mylibs/GSL/include
make
Note that you may need to give the same -DMSGSL_ROOT=~/mylibs/GSL/include hint
to CMake when configuring your own applications linking with Microsoft SEAL.
ZLIB is a widely used compression library that implements the DEFLATE compression
algorithm. Microsoft SEAL can use ZLIB (if present) to automatically compress data
that is serialized. For example, in some cases Ciphertext objects consist of a large
number of integers modulo specific prime numbers (coeff_modulus primes). When using
the CKKS scheme, although these prime numbers can often be quite small (e.g., 30 bits),
the numbers are nevertheless serialized as 64-bit integers. In this case, more than
half of data in a ciphertext are zeros that can be compressed away with a compression
library, such as ZLIB. The BFV scheme benefits typically less from this technique, because
the prime numbers used for the coeff_modulus encryption parameter tend to be larger,
and integers modulo these prime numbers fill more of each 64-bit word. The compression is
not only applied to Ciphertext objects, but to every serializable Microsoft SEAL object.
If ZLIB is detected by CMake, it will be automatically used for serialization (see
Serialization::compr_mode_default in native/src/seal/serialization.h. However, it is
always possible to explicitly pass compr_mode_type::none to serialization methods to
disable compression.
WARNING: The compression rate for a SecretKey can (in theory at least) reveal
information about the key. In most common applications of Microsoft SEAL the size of
a SecretKey would not be deliberately revealed to untrusted parties. If this is
a concern, one can always save the SecretKey in an uncompressed form by passing
compr_mode_type::none to SecretKey::save.
ZLIB is usually not found on a typical Windows system. You can clone it from
GitHub.com/madler/zlib to some convenient directory,
e.g., C:\MyLibs\zlib in this example. You need to build ZLIB first by opening
Developer Command Prompt for VS 2019, go to C:\MyLibs\zlib, and run
cmake .
cmake --build . --config Release
Next, you will need to signal Microsoft SEAL to enable ZLIB support by creating a new
Windows environment variable ZLIB_ROOT, and setting its value to C:\MyLibs\zlib.
Restart Visual Studio at this point if you had it open, otherwise it will not have
captured the newly created environment variable. Rebuilding Microsoft SEAL should now
automatically detect that ZLIB is available, and enable support for
compr_mode_type::deflate. To disable ZLIB support, delete the ZLIB_ROOT environment
variable, restart Visual Studio, and rebuild Microsoft SEAL.
The ZLIB (development package) can be conveniently obtained through a package manager
on most Linux distributions, e.g., on Ubuntu it suffices to install the package
zlib1g-dev. Alternatively, clone from GitHub.com/madler/zlib
and build it yourself. For example, suppose you have cloned ZLIB to ~/mylibs/zlib.
To build ZLIB, simply execute:
cd ~/mylibs/zlib
cmake .
make
If ZLIB was installed with a package manager, CMake will likely detect the location
of ZLIB automatically. Otherwise, if ZLIB was built in ~/mylibs/zlib, you need
to provide CMake with this location when building Microsoft SEAL as follows:
cd native/src
cmake . -DZLIB_ROOT=~/mylibs/zlib
make
Microsoft SEAL provides a .NET Standard library that wraps the functionality in Microsoft SEAL for use in .NET development.
The Microsoft Visual Studio 2019 solution file SEAL.sln contains the projects necessary
to build the .NET assembly, a backing native shared library, .NET examples, and unit
tests.
Microsoft SEAL for .NET requires a native library that is invoked by the managed .NET
library. Build the SEALNetNative project dotnet\native\SEALNetNative.vcxproj from
SEAL.sln. Building SEALNetNative results in the dynamic library sealnetnative.dll
to be created in dotnet\lib\$(Platform)\$(Configuration). This library is meant to
be used only by the .NET library, not by end users, and needs to be present in the same
directory as your executable when running a .NET application.
Once you have built the shared native library (see above), build the SEALNet project
dotnet\src\SEALNet.csproj from SEAL.sln. Building SEALNet results in the assembly
SEALNet.dll to be created in dotnet\lib\$(Configuration)\netstandard2.0. This
is the assembly you can reference in your application.
Build the SEALNetExamples project dotnet\examples\SEALNetExamples.csproj from
SEAL.sln. This results in the assembly SEALNetExamples.dll to be created in
dotnet\bin\$(Configuration)\netcoreapp2.1. The project takes care of copying the
native SEALNetNative library to the output directory.
Build the SEALNet Test project dotnet\tests\SEALNetTest.csproj from SEAL.sln. This
results in the SEALNetTest.dll assembly to be created in
dotnet\lib\$(Configuration)\netcoreapp2.1. The project takes care of copying the
native SEALNetNative library to the output directory.
To use Microsoft SEAL for .NET in your own application you need to:
- add a reference in your project to
SEALNet.dll; - ensure
sealnetnative.dllis available for your application when run. The easiest way to ensure this is to copysealnetnative.dllto the same directory where your application's executable is located.
Alternatively, you can build and use a NuGet package; see instructions in NUGET.md.
Microsoft SEAL for .NET relies on a native shared library that can be easily configured and built using CMake (>= 3.12) and a modern version of GNU G++ (>= 6.0) or Clang++ (>= 5.0). In macOS the Xcode toolchain (>= 9.3) will work.
For compiling .NET code you will need to install a .NET Core SDK (>= 2.1). You can follow these instructions for installing in Linux, or for installing in macOS.
If you only intend to run the examples and unit tests provided with Microsoft SEAL, you do not need to install the native shared library, you only need to compile it. The SEALNetExamples and SEALNetTest projects take care of copying the native shared library to the appropriate assembly output directory.
To compile the native shared library you will need to:
- Compile Microsoft SEAL as a static or shared library with Position-Independent Code (PIC);
- Compile native shared library.
The instructions for compiling Microsoft SEAL are similar to the instructions described, but
in addition you need to ensure that the CMake configuration option SEAL_LIB_BUILD_TYPE is
set to either Static_PIC (default) or Shared. Assuming Microsoft SEAL was built using
the default CMake configuration options, we can immediately use it to compile the shared
native library required for .NET:
cd dotnet/native
cmake .
make
cd ../..
To build the .NET Standard library, do the following:
cd dotnet/src
dotnet build
cd ../..
You can use the dotnet parameter --configuration <Debug|Release> to build either
a Debug or Release version of the assembly. This will result in a SEALNet.dll
assembly to be created in dotnet/lib/$(Configuration)/netstandard2.0. This assembly
is the one you will want to reference in your own projects.
To build and run the .NET examples, do:
cd dotnet/examples
dotnet run
cd ../..
As mentioned before, the .NET project will copy the shared native library to the assembly
output directory. You can use the dotnet parameter --configuration <Debug|Release> to
run either Debug or Release versions of the examples.
To build and run the .NET unit tests, do:
cd dotnet/tests
dotnet test
cd ../..
All unit tests should pass. You can use the dotnet parameter --configuration <Debug|Release>
to run Debug or Relase unit tests, and you can use --verbosity detailed to print the list
of unit tests that are being run.
To use Microsoft SEAL for .NET in your own application you need to:
- add a reference in your project to
SEALNet.dll; - ensure the native shared library is available for your application when run. The easiest way
to ensure this is to copy
libsealnetnative.soto the same directory where your application's executable is located.
In Linux or macOS, if you have root access to the system, you have the option to install the native shared library globally. Then your application will always be able to find and load it.
Assuming Microsoft SEAL is build and installed globally, you can install the shared native library globally as follows:
cd dotnet/native
cmake .
make
sudo make install
cd ../..
Using Microsoft SEAL will require the user to invest some time in learning fundamental
concepts in homomorphic encryption. The code comes with heavily commented examples that
are designed to gradually teach such concepts as well as to demonstrate much of the API.
The code examples are available (and identical) in C++ and C#, and are divided into
several source files in native/examples/ (C++) and dotnet/examples/ (C#), as follows:
| C++ | C# | Description |
|---|---|---|
examples.cpp |
Examples.cs |
The example runner application |
1_bfv_basics.cpp |
1_BFV_Basics.cs |
Encrypted modular arithmetic using the BFV scheme |
2_encoders.cpp |
2_Encoders.cs |
Encoding more complex data into Microsoft SEAL plaintext objects |
3_levels.cpp |
3_Levels.cs |
Introduces the concept of levels; prerequisite for using the CKKS scheme |
4_ckks_basics.cpp |
4_CKKS_Basics.cs |
Encrypted real number arithmetic using the CKKS scheme |
5_rotation.cpp |
5_Rotation.cs |
Performing cyclic rotations on encrypted vectors in the BFV and CKKS schemes |
6_performance.cpp |
6_Performance.cs |
Performance tests for Microsoft SEAL |
It is recommeded to read the comments and the code snippets along with command line printout from running an example. For easier navigation, command line printout provides the line number in the associated source file where the associated code snippets start.
WARNING: It is impossible to use Microsoft SEAL correctly without reading all examples or by simply re-using the code from examples. Any developer attempting to do so will inevitably produce code that is vulnerable, malfunctioning, or extremely slow.
This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.microsoft.com.
When you submit a pull request, a CLA-bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., label, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.
Pull requests must be submitted to the branch called contrib.
This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact [email protected] with any additional questions or comments.
For contributing to Microsoft SEAL, please see CONTRIBUTING.md.
To cite Microsoft SEAL in academic papers, please use the following BibTeX entries.
@misc{sealcrypto,
title = {{M}icrosoft {SEAL} (release 3.4)},
howpublished = {\url{https://github.com/Microsoft/SEAL}},
month = oct,
year = 2019,
note = {Microsoft Research, Redmond, WA.},
key = {SEAL}
}
@misc{sealcrypto,
title = {{M}icrosoft {SEAL} (release 3.3)},
howpublished = {\url{https://github.com/Microsoft/SEAL}},
month = june,
year = 2019,
note = {Microsoft Research, Redmond, WA.},
key = {SEAL}
}
@misc{sealcrypto,
title = {{M}icrosoft {SEAL} (release 3.2)},
howpublished = {\url{https://github.com/Microsoft/SEAL}},
month = feb,
year = 2019,
note = {Microsoft Research, Redmond, WA.},
key = {SEAL}
}
@misc{sealcrypto,
title = {{M}icrosoft {SEAL} (release 3.1)},
howpublished = {\url{https://github.com/Microsoft/SEAL}},
month = dec,
year = 2018,
note = {Microsoft Research, Redmond, WA.},
key = {SEAL}
}
@misc{sealcrypto,
title = {{M}icrosoft {SEAL} (release 3.0)},
howpublished = {\url{http://sealcrypto.org}},
month = oct,
year = 2018,
note = {Microsoft Research, Redmond, WA.},
key = {SEAL}
}