A Full-Stack Search Technique for Domain Optimized Deep Learning Accelerators
The rapidly-changing deep learning landscape presents a unique opportunity for building inference accelerators optimized for specific datacenter-scale workloads. We propose Full-stack Accelerator Search Technique (FAST), a hardware accelerator search framework that defines a broad optimization environment covering key design decisions within the hardware-software stack, including hardware datapath, software scheduling, and compiler passes such as operation fusion and tensor padding. In this paper, we analyze bottlenecks in state-of-the-art vision and natural language processing (NLP) models, including EfficientNet and BERT, and use FAST to design accelerators capable of addressing these bottlenecks. FAST-generated accelerators optimized for single workloads improve Perf/TDP by 3.7x on average across all benchmarks compared to TPU-v3. A FAST-generated accelerator optimized for serving a suite of workloads improves Perf/TDP by 2.4x on average compared to TPU-v3. Our return on investment analysis shows that FAST-generated accelerators can potentially be practical for moderate-sized datacenter deployments.
Code (0)
등록된 구현이 없습니다.
Tasks
Optical Character Recognition (OCR)SchedulingMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
CFU Playground: Full-Stack Open-Source Framework for Tiny Machine Learning (tinyML) Acceleration on FPGAs
Need for the efficient processing of neural networks has given rise to the development of hardware accelerators. The increased adoption of specialized hardware has highlighted the need for more agile design flows for har…
CPUArbor: Tree Search as a Cognition Layer for Autonomous Agents
Arbor is a multi-agent framework that introduces structured tree search as a cognition layer for autonomous agents operating in large, stateful action spaces. Prior autonomous optimization systems operate on isolated tar…
Try with Simpler -- An Evaluation of Improved Principal Component Analysis in Log-based Anomaly Detection
The rapid growth of deep learning (DL) has spurred interest in enhancing log-based anomaly detection. This approach aims to extract meaning from log events (log message templates) and develop advanced DL models for anoma…
Anomaly DetectionFullStack Bench: Evaluating LLMs as Full Stack Coders
As the capabilities of code large language models (LLMs) continue to expand, their applications across diverse code intelligence domains are rapidly increasing. However, most existing datasets only evaluate limited appli…
SERCNN: Stacked Embedding Recurrent Convolutional Neural Network in Depression Detection on Twitter
Conventional approach of self-reporting-based screening for depression is not scalable, expensive, and requires one to be fully aware of their mental health. Motivated by previous studies that demonstrated great potentia…
Depression Detection