Like other members of Intel® Performance Libraries, Intel Integrated Performance Primitives is a collection of high-performance code that performs domain-specific operations. It is distinguished by providing a low-level, stateless interface.
Based on experience in developing and using Intel Performance Libraries, Intel IPP has the following major distinctive features:
Intel IPP provides basic low-level functions for creating applications in several different domains, such as signal processing, image and video processing, operations on small matrices, and cryptography applications.
Intel IPP functions follow the same interface conventions, including uniform naming conventions and similar composition of prototypes for primitives that refer to different application domains.
Intel IPP functions use an abstraction level which is best suited to achieve superior performance figures by the application programs.
To speed up the performance, Intel IPP functions are optimized to use all benefits of Intel® architecture processors. Besides this, most of Intel IPP functions do not use complicated data structures, which helps reduce overall execution overhead.
Intel IPP is well-suited for cross-platform applications. For example, functions developed for the IA-32 architecture can be readily ported to the Intel® 64 architecture-based platform. In addition, each Intel IPP function has its reference code written in ANSI C, which clearly presents the algorithm used and provides for compatibility with different operating systems.
Product and Performance Information
Notice revision #20201201