Imagination launches multi-core IMG Series4 NNA

2 mins read

Imagination Technologies has announced the launch of the IMG Series4, its next-generation neural network accelerator (NNA) for advanced driver-assistance systems (ADAS) and autonomous driving.

The Series4 targets automotive industry-leading disruptors as well as Tier 1s, original equipment manufacturer (OEMS) and automotive semiconductor system-on-chip (SoC) manufacturers.

It features a new multi-core architecture, and the Series4 delivers ultra-high performance of 600 tera operations per second (TOPS) and beyond, offering low bandwidth and very low latency for large neural network workloads.

With the automotive industry developing new use cases such as self-driving cars and robotaxis demanding new levels of artificial intelligence (AI) performance, Imagination has already been working with leading industry players and innovators and other industries where functional safety is valued. As a consequence, the Series4 has already been licensed and will be available on the market in December 2020.

Imagination’s low-power NNA architecture is designed to run full network inferencing while meeting functional safety requirements. It executes multiple operations in a single pass to maximise performance per watt and deliver its industry-leading energy efficiency.

The Series4 includes:

  • Multi-core scalability and flexibility: Multi-core allows for flexible allocation and synchronisation of workloads across the cores. Imagination’s software, which provides fine-grained control and increases flexibility through batching, splitting and scheduling of multiple workloads, can now be exploited across any number of cores. Available in configurations of 2, 4, 6, or 8 cores per cluster.
  • Ultra-high performance: Series4 offers 12.5 TOPS per core at less than one watt. For example, an 8-cluster core can deliver 100 TOPS: thus, a 6x100 solution offers 600 TOPS. A Series4 NNA achieves performance that is over 20x faster than an embedded GPU and 1000x faster than an embedded CPU for AI inference.
  • Ultra-low latency: By combining all the cores into a 2, 4, 6 or 8-core cluster, all the cores can be dedicated to executing a single task, reducing latency, and therefore response time, by a corresponding factor. For example, for an 8-core cluster by a factor of eight.
  • Major bandwidth savings: Imagination’s Tensor Tiling (ITT), new to Series4 is a patent-pending technology that addresses bandwidth efficiency by splitting input data tensors into multiple tiles for efficient data processing. ITT exploits local data dependencies to keep intermediate data in on-chip memory. This minimises data transfers to external memory, reducing bandwidth by up to an incredible 90%. ITT is a scalable algorithm with major benefits on networks with large input data sizes.
  • Automotive safety: Series4 includes IP-level safety features and a design process that conforms to ISO 26262 to help customers to achieve certification. ISO 26262 is the industry safety standard that addresses risk in automotive electronics. Series4 enables the safe inference of a neural network without impacting performance. Hardware safety mechanisms protect the compiled network, the execution of the network and the data-processing pipeline.