33 C
Thursday, July 29, 2021
Home Technology Intel debuts Xe-based server GPU with focus on Android cloud gaming -...

Intel debuts Xe-based server GPU with focus on Android cloud gaming – Times of India

Intel has debuted its first discrete graphics processing unit (GPU) for the information heart, Intel Server GPU, based mostly on the Xe-LP microarchitecture and designed particularly for prime-density, low-latency Android cloud gaming and media streaming.
“Today is a key moment in our ambitious oneAPI and XPU journey. With the gold release of our oneAPI toolkits, we have extended the developer experience from familiar CPU programming libraries and tools to include our vector-matrix-spatial architectures. We are also launching our first data center GPU based on Xe-LP microarchitecture focused on the fast-growing cloud gaming and media streaming segments,” stated Raja Koduri, Intel senior vice chairman, chief architect and common supervisor of Architecture, Graphics and Software.
As the world strikes into an period of billions of clever units and an exponential progress of knowledge, CPUs alone require a shift in focus to a combination of architectures throughout CPUs, GPUs, FPGAs and different accelerators. Intel collectively describes this as ‘XPU’ imaginative and prescient. The launch of the Intel Server GPU is alleged to be one other step to increase the corporate’s providing within the XPU period.
With this, builders will be capable to entry a typical, open and requirements-based mostly programming mannequin throughout Intel XPUs with its oneAPI toolkits. “They deliver on the performance potential of the underlying hardware and lower software development and maintenance costs while reducing risks associated with deploying accelerated computing relative to proprietary, vendor-specific solutions,” stated the corporate in a press release.
Intel Server GPU relies on Xe-LP microarchitecture, its graphics structure, providing a low-energy, discrete system-on-chip design, with a 128-bit pipeline and 8GB of onboard low-energy DDR4 reminiscence. “By using Intel Server GPUs alongside Intel Xeon Scalable processors, service providers can scale graphics capacity separately from the server count, supporting larger numbers of streams and subscribers per system, while still achieving low TCO,” claims the corporate.
Packaging collectively 4 Intel Server GPUs in H3C’s three-quarter-size, full-peak x16 PCIe Gen 3.0 add-in card is claimed to assist greater than 100 simultaneous Android cloud gaming customers in a typical two-card system and as much as 160 simultaneous customers, relying on the precise recreation title and server configuration.
Intel is working with numerous software program and providers companions, together with Gamestrea
m, Tencent Games and Ubitus, to convey the Intel Server GPU to market. “Intel is an important collaborator on our Android Cloud Gaming solution. Intel Xeon Scalable processors and Intel Server GPUs offer a high-density, low-latency, low-power, low-TCO solution. We are able to generate over 100 game instances per two-card server for our most popular games, King of Glory and Arena of Valor,” stated Allen Fang, vice common supervisor of Tencent XianYou Cloud Gaming Platform.
The Xe-LP-based mostly Intel Server GPU is now transport. Along with the not too long ago launched Iris Xe MAX graphics, the GPU goals to enhance the visible computing expertise for customers all over the world.
Announced at SuperComputing 2019, the oneAPI trade initiative is alleged to be a imaginative and prescient for a unified and simplified cross-structure programming mannequin that delivers efficiency with out proprietary lock-in whereas enabling the mixing of legacy code. With oneAPI, builders can select the the structure for particular drawback that they’re making an attempt to unravel while not having to rewrite software program for the following structure and platform.
Intel introduced that the gold launch of oneAPI toolkits can be transport in December this 12 months. It can be obtainable free of charge, regionally and within the Intel DevCloud, alongside with industrial variations that embrace worldwide assist from the corporate’s technical consulting engineers. Intel will transition Parallel Studio XE and System Studio device suites to its oneAPI merchandise.
In addition, the Intel DevCloud, the place builders can take a look at code and workloads on a range of architectures, is increasing to incorporate new Iris Xe graphics {hardware}. Iris Xe MAX graphics is accessible for public entry; and Xe-HP to pick builders.
Intel Graphics Software: What’s new
One of Intel’s core methods scaling GPUs from entry-stage graphics to excessive efficiency computing (HPC) is the implementation of one code base. Toward this imaginative and prescient, Intel’s software program stack now claims to helps a number of generations of graphics, together with not too long ago-launched eleventh Gen Intel Core processors with built-in Iris Xe graphics and Iris Xe MAX discrete graphics. Expanding the code base to assist knowledge heart merchandise the place Linux is extra prevalent is subsequent step in its Xe structure technique.
The firm has created Project Flipfast to enhance the Linux gaming expertise. The Flipfast stack permits finish customers to run a graphical software in a digital machine (VM) whereas retaining native GPU efficiency and full host integration with zero-copy sharing between the VM and host. It claims to boost gaming efficiency, and the expertise straight applies to knowledge heart recreation streaming functions.
Intel stated that Implicit SPMD Program Compiler (ISPC) will run on high of oneAPI Level Zero, which offers low-stage, direct-to-metallic interfaces tailor-made to units within the oneAPI platform and is the general {hardware} abstraction layer. ISPC, powered by oneAPI, is a variant of the C programming language, which permits single program, a number of knowledge programming and is used to speed up Intel OSPRay ray tracing engine on Intel CPUs.

Source link

Most Popular