ARM’s DynamIQ processor architecture is oriented at AI
Tue 21 Mar 2017
ARM has revealed a new micro-architecture called DynamIQ, aimed at industry-scale artificial intelligence computing tasks, and addressing the legacy needs of manufacturers and providers who need to make saleable resources out of a wide range of existing CPU frameworks.
The company believes that the introduction of DynamIQ represents a ‘monumental shift’ in the evolution of multi-core microarchitecture, and for the possibilities it brings to ARM’s Cortex-A processor offerings.
DynamIQ offers a new core schema that permits variegated CPUs to co-exist in memory subsystem, which it claims improves latency and responsiveness for active applications in a multi-platform environment.
The new framework, as would be expected from this particular company, has fine-grained power management tools to help applications run in the most optimised manner possible. Users are able to control CPU speeds and can benefit from speedier power-switching modes (i.e. toggling between on/off/sleep). Additionally DynamIQ allows for partial memory sub-system power-downs; it’s a toolbox pre-armed against environmental concerns.
ARM’s product marketing head John Ronco commented:
“It’s a step change in how we build CPUs and the way we stitch CPUs together…It’ll be in smartphones and tablets, for sure, but also automotive networking and a whole range of other embedded devices. Anywhere a Cortex processor is used today, Dynamiq is going to be the next step forward.”
DynamIQ develops ARM’s big.LITTLE conception, wherein higher-level CPUs are paired with less energy-demanding, lower-yield chips. The advance is that DynamIQ is able to connect up to eight varying CPU types in any configuration, under the banner ‘heterogeneous computing’, effectively offering itself as a Rosetta stone for legacy architecture in the new AI gold rush.
DynamIQ uses the ancillary processors for all low-grade demands, activating the higher-powered CPUs only for those requests which require the extra computing power.
Ronco reports that DynamIQ is currently licensed to a number of vendors, with devices expected to hit the market early in 2018.
The new framework is extensible enough to allow dedicated solutions which need on-chip control of CPU groups, effectively allowing manufacturers to embed AI accelerators directly into their hardware.
Cortex-A processors under DynamIQ will be able scale multi-core configurations up to an 8-core limit, with fine control over the set-up of each processor, and configurability of power characteristics and performance profiles. It also brings new processor instructions which will lend new Cortex-A CPUs a fifty-fold increase in the performance of AI-related tasks over the next five years, relative to current benchmarks for Cortex-A73 systems.
DynamIQ’s safety features also offer higher responsiveness under ADAS, permitting the building of ASIL-D compliant technologies for failover scenarios.