Intel® Integrated Performance Primitives

Production-ready building blocks for cross-platform performance. Develop high-performance vision, signal, security, and storage applications with this multithreaded software library.

Give Intel your input on Intel® Integrated Performance Primitives (Intel® IPP) to help make improvements to meet your needs: 
 

Multimedia and Data Processing Performance

Intel IPP is an extensive library of ready-to-use, domain-specific functions that are highly optimized for diverse Intel architectures. Its royalty-free APIs help developers:

  • Take advantage of Single Instruction, Multiple Data (SIMD) instructions
  • Improve the performance of computation-intensive applications, including signal processing, data compression, video processing, and cryptography
  • Reduce cost and time to market for software development and maintenance

Release Notes

 

What's New

  • Intel IPP cryptography introduced VAES for Intel® Advanced Vector Extensions 2 (Intel® AVX2) optimizations for the AES-GCM algorithm and Intel® Advanced Vector Extensions 512 (Intel® AVX-512) optimizations for RSA algorithm that help users securely transmit data faster.
  • Faster performance is seen in the Intel IPP image domain with Intel AVX-512 and VNNI-based optimizations for RGB to XYZ color conversion. 
  • Faster performance seen with Intel AVX-512 optimizations for the Intel IPP signal-processing domain statistical function L2 norm. 

What's Updated

Cryptography ​

  • Eliminated the need for a redundant buffer, increased efficiency, ease of use for API and adoptability, and incorporated new bug fixes for CBC and CFB modes for the SM4 algorithm​
  • Expanded offerings with CCM and GCM modes, enabling the cryptographic multibuffer library to provide a greater performance benefit when compared to scalar implementation
  • Added support for the asymmetric cryptographic algorithm SM2, and for key exchange protocol and encryption and decryption APIs, and added other bug fixes and security enhancements
  • Added mitigation for a frequency throttling side-channel attack for the ECB, CMAC, and GCM modes of the AES algorithm

Image Processing ​

  • Fixed bugs in the planar YCbCr to RGB conversion in Intel IPP image-processing color-conversion functions​

Signal Processing ​

  • Extended optimization for multirate finite impulse response (FIR) filtering on Intel IPP signal processing functions​
  • Fast Fourier transform (FFT) and discrete Fourier transform (DFT) support for the FP16 data type on 4th generation Intel® Xeon® Scalable processors and newer

Data Compression ​

  • Optimizations for a lossless compression method, zlib v1.2.13, in Intel IPP data compression. These new optimizations help improve the quality and speed of compression and decompression in various data compression applications. ​
Download as Part of the Toolkit

Intel IPP is included as part of the Intel® oneAPI Base Toolkit (Base Kit), which is a core set of tools and libraries for developing high-performance, data-centric applications across diverse architectures.

Download the Stand-Alone Version

A stand-alone download of Intel IPP is available. You can download binaries from Intel or choose your preferred repository.

      

Runtime Version

Develop in the Cloud

Build and optimize oneAPI multiarchitecture applications using the latest Intel-optimized oneAPI and AI tools, and test your workloads across Intel® CPUs and GPUs. No hardware installations, software downloads, or configuration necessary.

   

For FPGA and People's Republic of China (PRC) Based Developers

Domains and Workloads

Image Processing

Take visual information and convert it into manageable, usable data for further analysis and decision-making. These image-processing applications use Intel IPP:

  • Healthcare (medical imaging)
  • Computer vision
  • E-commerce (visual search)
  • Digital surveillance
  • Biometric identification
  • Factory machine vision
  • Advanced driver assistance systems (ADAS) for autonomous driving
  • Printing and printers
  • Image recognition and enhancement
  • Remote equipment operation
  • Gesture recognition
  • Illegal image recognition
  • Optical correction

Data Compression

Reduce the number of bits needed to store or transmit data. Intel IPP highly optimizes these common compression standards:

  • Lempel-Ziv-Storer-Szymanski (LZSS)
  • LZ77 (zlib)
  • Lempel-Ziv-Oberhumer (LZO)
  • Bzip2
  • LZ4
  • ZFP

Achieve significant performance gains with plug and play functions on applications such as these:

  • Internet portal data center
  • Data storage centers
  • Databases
  • Enterprise data management

Signal Processing

Enable information generation, transformation, and interpretation. Pull meaning from broad sources of data, helping modern communications that include:

  • Voice recognition
  • Biotechnology
  • Wearable technology
  • Hearing aids
  • Speech synthesis

Optimize commonly used signal-processing functions for a wide variety of Intel architectures, including: 

  • Discrete Fourier transform (DFT)
  • Fast Fourier transforms (FFT)
  • Convolution
  • Filtering
  • Statistics

These signal-processing applications use Intel IPP:

  • Telecommunications
  • Energy
  • Ultrasound machines
  • Medical scanning
  • Record, enhance, and playback audio and non-audio signals
  • Echo cancellation: filter, equalize, and emphasis
  • Simulation of environment or acoustics
  • Games with sophisticated audio content or effect
  • Interfaces for voice-controlled personal assistants

Cryptography

Protect against cyberattacks and intrusion in the field of autonomous, self-driving cars with functions for: 

  • Security analysis
  • Threat intelligence
  • Mobile security
  • Cloud security
  • IoT security
  • Data integrity and authentication hash (SHA, MD5, SM3)
  • Public key cryptography (RSA, ECC, HMAC, CMAC)
  • Secure data transfer, such as:
    • Symmetric algorithms
    • Advanced Encryption Standard (AES)
    • Triple DES (TDES)
    • SMS4
    • Steam ciphers

The cryptography library is available as an open source library. GitHub*

These security applications use Intel IPP:

  • Telecommunications
  • Transaction security and cybersecurity
  • Smart card and wallet interfaces
  • ID verification
  • Copy protection
  • Electronic signature
  • Advanced driver-assistance systems (ADAS)

Benchmarks

These benchmarks illustrate the performance capabilities of Intel IPP.

Sample

Description
Multithread Image Resize

Learn how to use the ippiResize functionality in single and multithread modes.

Use the parallel_for loop functionality for external multithreading. The multithreading mode works if the project is built with Intel® oneAPI Threading Building Blocks (oneTBB) support.
Image Linear Transform (ipp_fft)

This example shows how to use two image linear transforms:

  • Fourier transforms (fast and discrete)
  • Discrete cosine transform (DCT)
External Threading Example (ipp_thread) Learn about threading for an image harmonization filter. Since internal threading is deprecated, it is important to know how to externally thread a generic Intel IPP function.
Cryptography Examples

Learn how to use the Intel IPP cryptography library with a collection of code samples on GitHub that cover:

  • Advanced Encryption Standard (AES)
  • Digital Signature Algorithm (DSA)
  • Asymmetric cryptography algorithm (RSA)
  • ShāngMì 4 block cipher algorithm (SM4)
View all Show less

🗗 Training & Events Calendar

Testimonials

Hisense* Adopts the Base Kit to Develop Its New Ultrasound Imaging Application

The Hisense* new ultrasound imaging platform consists of GPUs and CPUs in its computing architecture for ultrasound desktop and portable devices. They not only migrated key CUDA* kernel functions to SYCL* to run on current and future generations of Intel CPUs using the Intel® DPC++ Compatibility Tool, but also simplified programming for their cross-architecture systems. Using Intel® oneAPI Base Toolkit components including the Intel® oneAPI DPC++/C++ Compiler, Intel IPP, Intel® oneAPI Math Kernel Library (oneMKL), and oneTBB has helped in reducing the total post-processing time in the ultrasonic system for signals and images. Intel toolkits help them simplify their development efforts and deliver better products to their end users.

Johns Hopkins University School of Medicine

Optimization of processing and analysis of fMRI data as shown will help not only with operational efficiency for clinical fMRI but for large-scale research projects, which would benefit significantly from such speed increases while potentially reducing cloud deployment cost by approximately 25%. These optimizations were achieved by using Intel® C++ Compiler Classic, oneMKL, Intel® Performance Primitives library, and Intel® VTune™ Profiler.

Meituan*

"Intel® toolkits helped increase our end-to-end application processing performance on Intel® Xeon® platforms. By using oneAPI technology including Intel IPP, which is a set of high-performance software library combined with hardware deep optimization providing a large number of signal processing, image processing and other functions, we were able to significantly improve our image-processing code performance by 2.7x in image rotation, and 4x in image resize. This allows us to analyze larger image datasets, and builds the cutting-edge visual AI Inference solution for our end customers."

Specifications

Processors:
  • Intel® Core™ processors
  • Intel® Xeon® processors
  • Intel Atom® processors
Languages:
  • Built-in support for C and C++
Operating systems:
  • Windows
  • Linux
Development environments:
  • Compatible with compilers from vendors that follow platform standards, such as Microsoft*, GNU Compiler Collection (GCC)*, and Intel
  • Can be integrated with Microsoft Visual Studio*
License:

End User License Agreement (EULA)
Note Your installed software development products may use a prior version of the EULA.

Get Help

Your success is our success. Access these support resources when you need assistance.

Stay in the Know with All Things CODE

Sign up to receive the latest trends, tutorials, tools, training, and more to
help you write better code optimized for CPUs, GPUs, FPGAs, and other
accelerators—stand-alone or in any combination.