Intel On Technology Innovation Summit kicked off today, announcing the establishment of a unified developer zone and the launch of the 12th generation Of Intel Core Processors based On Intel 7 process, oneAPI 2022 toolkit and oneAPI Center of Excellence.
Intel's 12th generation Core processor announced several performance improvements
At the innovation Summit, Intel unveiled its 12th generation Core processor. It uses Intel 7nm process and high-performance hybrid architecture, including 60 processors and more than 500 model designs. The 12th generation Core processor product family is equipped with up to 16 cores and 24 threads, supporting fast performance core response and DDR5 memory ultra-fast data transfer,Photo editing performance is up 36% and video editing performance is up 32%. 3D modeling performance increased by 37% and multi-frame rendering speed increased by 100%.The core 12 also provides an overclocked tool that can overclocked energy efficient core and DDR5 memory using the latest Intel Ultimate debugging utility 7.5.
Intel has also released the first generation of Its slick, high-performance stand-alone GPU, code-named Alchemist, designed with gaming in mind. The Sleek Gpus offer Xe supersampling technology that game developers can incorporate into their games. At the same time, the cutting-edge GPU supports The Deep Link technology of Intel platform and Hyper Encode computing power. Intel says it can realize the simultaneous acceleration of single video file transcoding across integrated graphics card and independent graphics card engine.
In addition, Intel also announced the next generation Intel Xeon scalable processor code-named Sapphire Rapids, based on enhanced SuperFin technology, including DDR5, PCIe Gen 5, Compute Express Link 1.1 and other technology standards. Its optimization can solve the cost problem in the large-scale deployment mode of data center, improve the utilization of processor kernel, and reduce the cost of power consumption and floor space. Intel plans to achieve a 30-fold increase in overall AI performance gains over the previous generation with the next generation of Intel Xeon scalable processors.
Launched the first Intel-specific ASIC IPU in cooperation with Google Cloud
Intel also announced deep cooperation with Google Cloud to launch Intel's first dedicated ASIC IPU code-named Mount Evans. The solution is supported by industry standard programming languages and open source infrastructure programmer development kits. Intel also unveiled Intelligent Fabirc, an end-to-end programmable platform that leveragesIntel's hardware and software products to create business opportunities and put control in the hands of developers.
Integrate developer directories to create a unified developer zone
Wang Rui, chairman of Intel China, said Intel will return with innovative leadership in technology and products, announcing that it will once again lead the industry in performance in 2025 and implement Z-class computing by 2027. Wang also said Intel will co-create an open ecosystem and make key investments for developers, including a developer zone and oneAPI Center of Excellence, to ensure developers have optimized tools and software environments to accelerate their deployment.
Intel CEO Pat Middot; Developers are the true "superheroes" of a digital world powered by semiconductors, and Intel will continue to empower developers with the power of silicon to usher in a new era of innovation, Kissinger said.
At the conference, Intel announced that it will launch an integrated Intel developer catalog containing key Intel software products, as well as an optimized Intel DevCloud development environment for testing and running workloads and software tools on a wide range of Intel's latest hardware.
Intel detailed the oneAPI 2022 toolkit, which offers over 900 new features that for the first time add cross-architecture software development capabilities to cpus and gpus through a unified C++/SYCL/Fortran compiler and Data Parallel Python. In addition, the oneAPI foundation toolkit adds compilers, profilers, debuggers, and migration tools to the canonical language and library, and provides cross-architecture compatibility that increases developer productivity and innovation, effectively avoiding the financial and technical burdens of today's proprietary programming models.
Intel predicts that by 2030, everyone could have petaflops of computing power and a terabyte of data, with a delay of less than a millisecond.
Supercomputers will be powered by next-generation Intel processors
Finally, Pat & Middot; The Argonne National Laboratory's Aurora supercomputer, powered by Intel's next-generation Xeon scalable processor and Intel's next-generation GPU code-named Ponte Vecchio, delivers double precision peak computing performance of more than two trillion peak-calculations per second, Kissinger announced. The Aurora system will have exaFLOP performance, the equivalent of a petabyte of floating point calculations per second, and will be able to handle both traditional high-performance computing and artificial intelligence.