Case Study: Tencent Optimizes Image Identification

Submitted: June 27, 2016 Last updated: June 27, 2016
  • File:
    tencent-case-study.pdf
  • Size:
    1.04 MB
Download

Detailed Description

Tencent, Inc., is China’s largest and most-used Internet service portal. It owns both the largest online game community and the largest Web portal (qq.com), as well as the No. 1 and No. 2 applications (WeChat*, QQ*) in China.

Every day, Tencent needs to process billions of new user-generated images fromWeChat, QQ, and QQ Album*. Some hot applications even have hundreds of millions of images to be uploaded, stored, processed, and downloaded in a single day―which consumes vast computing resources.

To manage, store, and process these images, Tencent developed Tencent File System* (TFS*). But even with compression, the image volume reached hundreds of petabytes. Moreover, it is still growing explosively―and the supported cluster has more than 20,000 servers.

Originally, Tencent used the md5sum* utility tool along with the Operator* OS to compute the MD5 value for each image file. Intel worked closely with Tencent engineers to help them optimize performance with Intel® Integrated Performance Primitives (Intel® IPP), which helped Tencent achieve a 100 percent performance improvement on the Intel® architecture-based platform.

Product and Performance Information

1

Intel's compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.

Notice revision #20110804