CMOS+X: Stacking Persistent Embedded Memories based on Oxide Transistors upon GPGPU Platforms

TL;DR

This paper proposes integrating oxide transistor-based persistent memory in GPGPU architectures to overcome SRAM scaling limits, resulting in significant improvements in memory capacity, energy efficiency, and overall GPU performance.

Contribution

It introduces CMOS+X oxide-based memory cells as a novel solution for high-bandwidth GPGPU memories, demonstrating their density, energy benefits, and performance gains through detailed analysis and GPU benchmarking.

Findings

3x more read ports with 70% lower standby power

Up to 5.2x performance per watt improvement

8% higher geometric mean IPC

Abstract

In contemporary general-purpose graphics processing units (GPGPUs), the continued increase in raw arithmetic throughput is constrained by the capabilities of the register file (single-cycle) and last-level cache (high bandwidth), which require the delivery of operands at a cadence demanded by wide single-instruction multiple-data (SIMD) lanes. Enhancing the capacity, density, or bandwidth of these memories can unlock substantial performance gains; however, the recent stagnation of SRAM bit-cell scaling leads to inequivalent losses in compute density. To address the challenges posed by SRAM's scaling and leakage power consumption, this paper explores the potential CMOS+X integration of amorphous oxide semiconductor (AOS) transistors in capacitive, persistent memory topologies (e.g., 1T1C eDRAM, 2T0C/3T0C Gain Cell) as alternative cells in multi-ported and high-bandwidth banked GPGPU memories. A detailed study of the density and energy tradeoffs of back-end-of-line (BEOL) integrated memories utilizing monolithic 3D (M3D)-integrated multiplexed arrays is conducted, while accounting for the macro-level limitations of integrating AOS candidate structures proposed by the device community (an aspect often overlooked in prior work). By exploiting the short lifetime of register operands, we propose a multi-ported AOS gain-cell capable of delivering 3x the read ports in ~76% of the footprint of SRAM with over 70% lower standby power, enabling enhancements to compute capacity, such as larger warp sizes or processor counts. Benchmarks run on a validated NVIDIA Ampere-class GPU model, using a modified version of Accel-Sim, demonstrate improvements of up to 5.2x the performance per watt and an average 8% higher geometric mean instruction per cycle (IPC) on various compute- and memory-bound tasks.