Intel Rivals NVIDIA & AMD As It Introduces Its First GPU For Data Centers

Intel Server GPU

Chipmaker Intel unveiled its first discrete GPU for data centres. Named Intel Server GPU, it is designed to power Android-based cloud gaming and media streaming segments. A low-power chip, Intel Server GPU will be used in the new PCIe graphics card which can be placed inside the servers. Interestingly, Raja Koduri, Intel Senior Vice President had released a teaser picture of a ‘mysterious GPU’ on his Twitter account in May, leading many to believe a future announcement related to GPU aimed at the data centre.

The graphics card market has been largely dominated by AMD and Nvidia. However, with the new launch, Intel has made inroads. Notably, Intel already has launched Iris Xe Max, a discrete GPU for laptops.

During the announcement of Intel Server GPU, the company spokesperson also said that they would be releasing Intel oneAPI toolkit, to include vector-matrix-spatial architectures in developer experience, in December, along with new capabilities in the software stack as part of the company’s combined hardware and software design approach.


Sign up for your weekly dose of what's up in emerging technology.

Intel Server GPU

With the inclusion of Intel Server GPU, Intel aims at expanding its existing suite of platform innovations for improved gaming and media experiences. It is based on the Xe-LP microarchitecture, arguably the most energy-efficient graphics architecture from Intel. When used along with Intel Xeon Scalable processors and open-sourced licensed software, Intel Server GPU offers high-density, low-cost solutions for Android cloud gaming and video streaming. Further, with this arrangement, service providers can scale the graphics separately from server count that supports the number of streams and subscribers per system.

Intel is currently working on bringing various service partners including Gamestream, Tencent Games, and Ubitus for a market release. 

Intel has also introduced an enterprise-grade graphics card XG310, built specifically for the purpose of cloud streaming. It can accommodate four Intel Server GPUs, with each having a base clock speed of 0.9GHz and a boost clock speed of 1.1 GHz. It can run on low power costs, even less than counterpart Nvidia RTX 3080 card, which helps clients consolidate their server into smaller footprints.

Intel oneAPI Toolkit

The oneAPI industry initiative was first introduced at the SuperComputing 2019 event. It allows for a simplified cross-architecture programming model that performs without proprietary lock-in and enables the integration of legacy code. It eliminates the need for rewriting the software for the subsequent architecture and platform.

Intel oneAPI toolkits use hardware capabilities and instructions such as Intel AVX-512 and Intel DL Boost on CPUS, along with XPU features. With the latest announcement for the gold release of Intel oneAPI toolkit, it will now be made freely available locally and on the Intel DevCloud and other commercial versions.

Along with the endorsement from industry giants such as Microsoft Azure and TensorFlow, the University of Illinois Beckmann has already established a new oneAPI centre of excellence (CoE) for life sciences applications in additional computing environments. There are oneAPI CoEs in other universities such as the University of Stockholm and Heidelberg.

Updation on Intel Graphics Software

As per the new update, Intel’s software stack will now support multiple generations of graphics, including 11th Gen Intel Core processors with integrated Intel Iris Xe graphics and Intel Iris Xe MAX discrete graphics. Intel has enabled code reuse between operating systems by optimising the Linux driver. Further, with an increased focus on Linux 3D performance, Intel will now be delivering three fully integrated distribution-ready stacks.

Intel has also announced Intel Implicit SPMD Program Compiler (ISPC) that runs on top of oneAPI Level Zero. ISPC is the overall hardware abstraction layer that promises to provide a low-level and direct-to-metal interface for devices in the oneAPI platform. ISPC is basically a variant of the C programming language.

Notably, Intel, recently also introduced Project Flipfast for improved Linux gaming experience by allowing the end-users to run graphical applications in virtual machines without hampering native GPU performance and ensuring full host integration with no sharing between the virtual machine and the host. The advantage of Project Flipfast is that it increases gaming performance drastically.

More Great AIM Stories

Shraddha Goled
I am a technology journalist with AIM. I write stories focused on the AI landscape in India and around the world with a special interest in analysing its long term impact on individuals and societies. Reach out to me at

Our Upcoming Events

Conference, in-person (Bangalore)
Machine Learning Developers Summit (MLDS) 2023
19-20th Jan, 2023

Conference, in-person (Bangalore)
Rising 2023 | Women in Tech Conference
16-17th Mar, 2023

Conference, in-person (Bangalore)
Data Engineering Summit (DES) 2023
27-28th Apr, 2023

Conference, in-person (Bangalore)
MachineCon 2023
23rd Jun, 2023

Conference, in-person (Bangalore)
Cypher 2023
20-22nd Sep, 2023

3 Ways to Join our Community

Whatsapp group

Discover special offers, top stories, upcoming events, and more.

Discord Server

Stay Connected with a larger ecosystem of data science and ML Professionals

Subscribe to our newsletter

Get the latest updates from AIM