Intel® Integrated Performance Primitives (Intel® IPP) & Their Applications

Speed performance for imaging, vision, signal, security, and storage applications.

  • Multicore, multiple operating system, and multiplatform ready, computationally intensive and highly optimized functions.
  • Plug in and use APIs to quickly improve application performance.
  • Reduced cost and time-to-market (TTM) on software development and maintenance.
  • Take advantage of Priority Support―connect privately with Intel engineers for technical questions.

Case Studies

Tencent doubles the speed of its image filter system which detects illegal images by using Intel® Integrated Performance Primitives (Intel® IPP)

In popular apps like WeChat*, QQ*, and QQ Album* the volume of newly generated images reach about 100 petabytes. Some users may try to upload illegal images. The system has to run a check on each image to try to block them. Imagine trying to search through 100 petabytes of data! Intel® IPP filter function (ipp_filter2D) took 9ms to perform some imaging operations when compared to 143ms with OpenCV. The Intel® IPP filter2D is fifteen times faster than the OpenCV* plain code. Read this case study here.

JD.com speeds image processing 17 times with Intel® IPP

JD.com processes more online transactions than any other company in China, with a market share of 54.3 percent in the second quarter of 2014. The company's business has grown rapidly, from offering approximately 1.5 million SKUs in 2011 to approximately 25.7 million in 2013. Today, JD.com must handle petabytes of data, which takes an efficient, robust, distributed file system. JD.com speeds up its image processing seventeen times – handling 300,000 images in 162 seconds instead of 2800 seconds. Read this case study here.

Tencent doubles MD5 identification using Intel® IPP

Every day, Tencent needs to process billions of new user-generated images from WeChat, QQ, and QQ Album. Hundreds of millions of images needs to be uploaded, stored, processed, and downloaded in a single day―which consumes vast computing resources. Even with compression, the image volume reached hundreds of petabytes and is still growing explosively―and the supported cluster has more than 20,000 servers. Tencent engineers implemented Intel IPP MD5 for their online system. Their test showed about a 60 percent performance improvement compared to the original lMD5. Read this case study here.