High-Density on-chip memory

Drop-in SRAM Replacement

Important industry growth drivers, such as AI, ML, Automotive, etc. operate on ever-growing amounts of data that is typically stored off-chip in an external DRAM. Unfortunately, off-chip memory accesses are up-to 1000x more costly in latency and power compared to on-chip data movement. This limits the bandwidth and power efficiency of modern systems. In order to reduce these off-chip data movements, almost all SoCs incorporate large amounts of on-chip embedded memory caches that are typically implemented with SRAM and often constitute over 50% of the silicon area. This memory bottleneck is further aggravated since SRAM scaling has reached its limits and no longer shrinks in advanced CMOS process nodes beyond 5nm.

RAAAM™’s patented GCRAM technology provides up-to 50% area reduction and reduced power consumption by X3 compared to high-density SRAM by utilizing a unique bitcell solution. This enables increasing the on-chip memory capacity using the same die size to provide a dramatic improvement in system bandwidth and power efficiency through reduction or complete removal of off-chip data movement. Alternatively, reducing the memory footprint can lead to substantial fabrication cost savings through die size reduction.

In addition to its area benefits, the GCRAM bit-cell utilizes decoupled write and read ports, providing native two-ported operation at no additional cost, offering a substantial increase in memory bandwidth. Furthermore, GCRAM enables robust voltage scaling capabilities of down to 450mV read and write voltage in 16nm FinFET technology, delivering substantial power savings compared to the foundry and other SRAM solutions.

RAAAM™’s GCRAM technology has already been validated on silicon of leading semiconductor foundries in process nodes ranging from 180nm  – 5nm and successfully taped-out in 2nm technology.

• Up-to 2X Higher density vs. SRAM
• Up-to  3X power reduction vs. SRAM
• Standard SRAM interface
• Extended interface options vs. SRAM
• Single/two ported
• Up to 2Mbit instances
• Standard CMOS process
• Single cycle operation
• Customizable

• AI and ML – Weight / Input buffers
• IoT and MCU – System buffers
• Automotive – Large memory caches
• Media Processing – Image / Video buffers