Amazon says arm graviton chips are more efficient than x86
Table of contents:
We talked about the ARM Graviton chips a few weeks ago, where he claims that it was an AMD Opteron-based processor, anyway, Amazon seems to be very happy with the implementation of these processors on a large scale, and claims that they are much more efficient than any other x86 processor from AMD or Intel.
Graviton will allow savings of up to 45% in the costs of Amazon cloud services
The e-commerce giant says that the implementation of Graviton will lead to savings of up to 45% in the costs of its cloud services.
The ARM Graviton processor contains 64-bit Neoverse cores based on the 16nm Cosmos architecture. According to EE News Europe, Drew Henry, ARM senior vice president, said the Israeli-designed Graviton operates on the 64-bit Cortex-A72 core, which operates at clock rates up to 2.3GHz.
This does not mean that Amazon is going to abandon all its equipment based on Intel and AMD, but that ARM's Graviton chips will help to scale applications. Here users of any Amazon service may share the load across a group of smaller instances, such as containerized microservices, web servers, development environments, and caching fleets.
It also means that Amazon will now have the ability to license ARM blueprints, through Annapurna. Additionally, the company can customize and fine-tune those designs, as well as the ability to hire manufacturers like TSMC and Global Foundries to make competitive chips.
AWS is also building a custom ASIC for AI Inference, called Inferentia, for Amazon. This could be able to scale from hundreds to billions of operations per second and further reduce the cost of cloud-based operations. This will allow Amazon to be more competitive without raising costs too much compared to its rivals in the cloud computing space, such as Microsoft or Google.
Fudzilla fontThe radeon r9 nano is 50% more efficient than fury x
The radeon R9 Nano is 50% more energy efficient than the Fury X offering the performance of a 290X while consuming half
The iphone 6s with a processor manufactured by tsmc are more efficient with the use of the battery
IPhone 6S smartphones with Apple A9 processor manufactured by TSMC are more efficient with the use of battery
Nvidia comments that turing at 12nm is more efficient than vega at 7nm
Turing uses a 12nm node and is more efficient than AMD at 14nm (Vega 10 = Radeon RX Vega 64) and 7nm (Vega 20 = Radeon VII).