Arm is rebranding its system-on-a-chip product designs to showcase power savings for AI workloads, targeting a surprising sector

The Chip designer lever located in the UK offers architecture for AAAAA-CHIP (SOCS) used by the world’s largest technological brandsNVIDIA to Amazon to the Google parent company alphabet and already already Ever produces any device – though It was reported to be changed this year.

And you think with a The total revenues of the last quarter are $ 1.24 billion recordYou may simply want to keep things resistant and catch the cash.

However, the goal wants to be a part of the action with some customers who offer their record revenues by offering the AI ​​graphics processing units in the EU’s enterprise and some customers who unite their technology.

Today, The company has announced a new product naming strategy This emphasizes that the component is transferred to a platform-first company from a supplier of IP.

“Customers belong to customers that we offer more than just one device and chip design. Specifically – An entity General Marketing Officer Ami Badani, we have a wonderful ecosystem in an exclusive interview in connection with yesterday.

Indeed, the CEO of ARM reported to Rene Haas Tech News Outlet The next platform In February, as well as creating low energy chips than competition, the creation of head-down chips (cough cough, Intel) power-AC AI training and results were found well to serve as the basis of the work.

According to this article, today’s information center consumes about 460 teravatt clock electricity a year, but this decade is expected to increase by the end of the world, and 4 percent of the energy-saving chips are used by 4 percent of the world and used in infrastructure for these centers.

To the platform from IP: A significant change

AI workload reorganizes its proposals around the scale of complexity and power requirements, and the lever around full computing platforms.

These platforms allow faster integration, more efficient scaling and lower complexity for partners.

To reflect this change, the goal is scholarships in advance conventions and promote new product families organized by the market:

• Neovsse for infrastructure
• NIVA for PC
• Lumex for mobile
• Zena for the car
• Orbis for Iot and Edge AI

The Mali brand will continue to represent GPU victims, integrated in the form of components on these new platforms.

Along with the change in the name, the arm justifies the product numbering system. IP identifiers will now be adapted with Ultra, Rewards, Pro, Nano and Pico-labeled platforms and performance performance performance. This structure aims to make the road map more transparent to customers and developers.

Exploded with strong results

Rebranding is followed by a strong Q4 fiscal year, which is the strongest Q4 fiscal year, where the company for the first time in quarterly income.

The total revenue is 1.24 billion dollars, both record license revenues ($ 634 million, 53%) and royalties (607 million, 18%).

It should be noted that this fee was managed by ARMV9 architecture and arm reporting systems (CSS), smartphones, cloud infrastructure and Edge AI.

The mobile market was a parking lot: global smartphone shipments increased by less than 2%, the goal of the smartphone royalty increased by about 30%.

The company also entered the first car CSS contract with the leading AV manufacturer.

Alm, although the house did not open a precise name, Badani Venturebeat said that in addition to AI model providers and cloud hyperscalers such as Google and Amazon, the AI ​​said it was seen as a large growth area of ​​the car.

“We are looking at cars like a great growth area and believe that the AI ​​and other advances will be standard for our designs to be perfect.”

Meanwhile, cloud providers like AWS, Google Cloud and Microsoft Azure confirm the growing effect on the score center using the score to manage AI workloads to manage AI workloads.

The software and vertical integrated products have grown a new platform ecosystem

ARM complements the hardware platforms with expanded software tools and ecosystem support.

GitHub allows GITHUB to Copilot, allows users to optimize the code using the arm architecture of users.

More than 22 million developed goals are built on more than 22 million and its Kleidi’s AI program has exceeded 8 billion aggregate installs along the devices.

The ARM management sees the rebrand as a natural step in the long-term strategy. By providing performance and naming of vertical integrated platforms, the company aims to meet the growing demand compared to the Energy Efficient AI account from the device.

Like Haas wrote the goal in the blog postARM’s computing platforms are based on a future where the AI ​​is everywhere and the goal is preparing to surrender on the scale of this foundation.

What does AI and information mean for decisions

It is likely to change how techniques, AI, information and security roles are likely to approach the daily work and future planning.

A more clear platform structure for managers of a large language model is a more easy way to select an optimized counting architecture for the AI ​​workload.

As model placement time increases, the pre-designated computing systems such as essence and efficiency are increased, the pre-designated calculation systems can be reduced to the surface required to evaluate raw IP blocks and faster in iterative development periods.

Engineers, AI pipelines in environments, models and performance performance performance within the new ARM architecture can help standardize the pipeline.

This provides a practical way to change the calculation abilities with many job loading requirements – this is not to be managed in the restoration or resource intensive training work in the cloud.

These engineers can often find more clarity to map the system working time and expense performance, to map the logic of its orchestra to the predefined ARM platform platforms.

Information infrastructure leaders can also benefit from maintaining high transmission pipelines and ensure data integrity.

Naming updates and system level integration, signaled to a deeper commitment to support expandable designs that work well with AI effective pipelines.

Computing Substances can also accelerate the time-market for special silicone, which supports the next gene data platforms for budgeting, budget restrictions and the limited engineering group.

Meanwhile, security leaders will probably see the effects of security features and system-level compatibility on these platforms.

The goal, which aims to offer consistent architectures along the cloud, security teams, especially both performance and serious access controls, can be planned for the ending, planning and applying more easily to the end.

The wider effect of this brand change is a signal for enterprise architects and engineers