OneDSE: A Unified Microprocessor Metric Prediction and Design Space Exploration Framework

TL;DR

OneDSE introduces a unified framework combining a transformer-based workload-aware prediction model and an efficient optimizer to accelerate and improve the accuracy of CPU design space exploration under power and performance constraints.

Contribution

The paper presents OneDSE, a novel unified framework with a transformer-based prediction model and a reinforcement learning optimizer for faster, more accurate CPU design space exploration.

Findings

TrACE-p outperforms SOTA IPC prediction by 5.71x (single workload) and 28x (multiple workloads).

MAST outperforms SOTA metaheuristics by 1.19x and is an order of magnitude faster.

SMART-TrACE reduces prediction error by 10.6% over TrACE.

Abstract

With the slowing of Moores Law and increasing impact of power constraints, processor designs rely on architectural innovation to achieve differentiating performance. However, the innovation complexity has simultaneously increased the design space of modern high performance processors. Specifically, we identify two key challenges in prior Design Space Exploration (DSE) approaches for modern CPU design - (a) cost model (prediction method) is either slow or microarchitecture-specific or workload-specific and single model is inefficient to learn the whole design space (b) optimization (exploration method) is slow and inaccurate in the large CPU parameter space. This work presents a novel solution called OneDSE to address these emerging challenges in modern CPU design. OneDSE is a unified cost model (metric predictor) and optimizer (CPU parameter explorer) with three key techniques - 1. Transformer-based workload-Aware CPU Estimation (TrACE) framework to predict metrics in the parameter space (TrACE-p) and parameters in the in the metric space (TrACE-m). TrACE-p outperforms State of The Art (SOTA) IPC prediction methods by 5.71x and 28x for single and multiple workloads respectively while being two orders of magnitude faster. 2. We also propose a novel Metric spAce Search opTimizer (MAST) that leverages TrACE-m and outperforms SoTA metaheuristics by 1.19x while being an order of magnitude faster. 3. We propose Subsystem-based Multi-Agent Reinforcement-learning based fine-Tuning (SMART)-TrACE that achieves a 10.6% reduction in prediction error compared to TrACE, enabling more accurate and efficient exploration of the CPU design space.