ARM’s DynamIQ eyes performance for AI

Article By : Vivek Nanda

ARM targets the artificial intelligence market with new technology to boost multi-core flexibility and performance.

Seeing a growing demand for artificial intelligence (AI) products that enable automated systems—in driving, manufacturing and consumer products–ARM is developing its own answer in what they are calling "DynamIQ" technology.

Essentially the next step for Cortex-A processors, DynamIQ touts a multi-core microarchitecture. ARM has announced that the introduction of the technology is an evolutionary step forward for ARM big.LITTLE technology.

It carries on the ‘right processor for the right task’ approach from ARM and enables configurations of big and LITTLE processors on a single compute cluster, which was previously not possible. ARM's announcement provides an example: 1+3 or 1+7 DynamIQ big.LITTLE configurations with substantially more granular and optimal control are now possible.

ARM has provided a list of DynamIQ features that they will apply in new Cortex-A processors later this year.

  • Dedicated processor instructions for ML and AI: Cortex-A processors designed for DynamIQ technology can be optimised to deliver up to a 50x boost in AI performance over the next 3-5 years relative to Cortex-A73-based systems.
  • Increased multi-core flexibility: You can scale up to 8 cores in a single cluster and each core can have different performance and power characteristics.
  • Higher performance within restricted thermal budgets: ARM is promising faster switching of software tasks to match the "right-sized processor."
  • Safer autonomous systems: ARM is claiming greater levels of responsiveness with DynamIQ.

While the company is already making noise about the underlying technology, the first processors–possibly a few Cortex-A variants as earlier mentioned–are at least a few months away.

Leave a comment