The browser version you are using is not recommended for this site.Please consider upgrading to the latest version of your browser by clicking one of the following links.
We are sorry, This PDF is available in download format only
Company Established in 2007, Face.com* offers a unique cloud-based image processing service that can quickly and automatically identify people’s faces in photos. Face.com licenses the technology to several popular social networking sites to incorporate it into their infrastructure and offering to end users, as well as making it available to software developers through an application program interface (API). It also provides the technology to smartphone users through its KLIK* app, available on the iPhone*. When users take a picture, the app transmits the image data to Face.com’s servers, which perform the calculations needed to identify who is in the shot in a matter of seconds. Challenge To provide real-time face recognition, Face.com’s service relies on the processing capacity of servers within Face.com’s own data center, or those of the customers the service is licensed to, to analyze large volumes of image data on a pixel-by-pixel basis. To preserve the speed and quality of the service, it needs reliable and responsive processing support. Within its own operations, Face.com expects significant growth in the volume of traffic to its servers once its KLIK app is officially launched. To cope with this increased demand, it needs the processing capacity available to scale up the service. Solution As a long-term user of Intel® processors, Face.com saw the forthcoming arrival of the Intel® Xeon® processor E5 family and Intel® Advanced Vector Extensions (Intel® AVX) instruction set as an opportunity to boost the performance of its service for its licensed customers and its KLIK app. A trial was set up to compare the performance of the new processor with that of existing models. Working with Intel’s developer support team, Face.com optimized its face recognition processes to work effectively with the new hardware. In particular, it focused on modifying its code to harness the additional parallel bandwidth support Intel AVX provides for processor-intensive floating point calculations, which are a key part of its image analysis technology. Read the full Face.com Upgrades to Intel® Xeon® Processor E5 and Intel® AVX success brief.
A video overview of Intel® Datacenter Manager