Memory-centric Hardware Acceleration for Machine Intelligence

Even the most advanced AI chip architectures suffer from performance and energy efficiency limitations caused by the memory bottleneck between computing cores and data. Most state-of-the-art CPUs, GPUs, TPUs and other neural network hardware accelerators are limited by the latency, bandwidth and energy consumed to access data through multiple layers of power-hungry and expensive on-chip caches and external DRAMs. Near-memory computing, based on emerging nonvolatile memory technologies, enables a new range of performance and energy efficiency for machine intelligence. In this presentation, we introduce innovative and affordable near-memory processing architectures for computer vision and voice recognition, and also present architectural recommendations for edge computing and cloud servers. We also discuss how non-volatile memory technologies, such as Crossbar Inc.’s ReRAM, can be directly integrated on-chip with dedicated processing cores, enabling new memory-centric computing architectures. The superior characteristics of ReRAM over legacy non-volatile memory technologies help to address the performance and energy efficiency demands of machine intelligence at the edge and in the data center.