PerfSeer: An Efficient and Accurate Deep Learning Models Performance Predictor

TL;DR

PerfSeer introduces a graph neural network-based predictor for deep learning model performance, leveraging comprehensive model representations and novel optimizations to achieve high accuracy in predicting metrics like execution time and resource use.

Contribution

The paper presents PerfSeer, a novel GNN-based performance predictor that effectively captures model topology and features, outperforming existing methods in accuracy.

Findings

PerfSeer outperforms nn-Meter, Brp-NAS, and DIPPM in accuracy.

The dataset includes over 53,000 model configurations with various performance metrics.

SeerNet can predict multiple metrics simultaneously with minimal accuracy loss.

Abstract

Predicting the performance of deep learning (DL) models, such as execution time and resource utilization, is crucial for Neural Architecture Search (NAS), DL cluster schedulers, and other technologies that advance deep learning. The representation of a model is the foundation for its performance prediction. However, existing methods cannot comprehensively represent diverse model configurations, resulting in unsatisfactory accuracy. To address this, we represent a model as a graph that includes the topology, along with the node, edge, and global features, all of which are crucial for effectively capturing the performance of the model. Based on this representation, we propose PerfSeer, a novel predictor that uses a Graph Neural Network (GNN)-based performance prediction model, SeerNet. SeerNet fully leverages the topology and various features, while incorporating optimizations such as Synergistic Max-Mean aggregation (SynMM) and Global-Node Perspective Boost (GNPB) to capture the critical performance information more effectively, enabling it to predict the performance of models accurately. Furthermore, SeerNet can be extended to SeerNet-Multi by using Project Conflicting Gradients (PCGrad), enabling efficient simultaneous prediction of multiple performance metrics without significantly affecting accuracy. We constructed a dataset containing performance metrics for 53k+ model configurations, including execution time, memory usage, and Streaming Multiprocessor (SM) utilization during both training and inference. The evaluation results show that PerfSeer outperforms nn-Meter, Brp-NAS, and DIPPM.