IANUS: Integrated Accelerator based on NPU-PIM Unified Memory System

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Seo, Minseok, Nguyen, Xuan Truong, Hwang, Seok Joong, Kwon, Yongkee, Kim, Guhyun, Park, Chanwook, Kim, Ilkon, Park, Jaehan, Kim, Jeongbin, Shin, Woojae, Won, Jongsoon, Choi, Haerang, Kim, Kyuyoung, Kwon, Daehan, Jeong, Chunseok, Lee, Sangheon, Choi, Yongseok, Byun, Wooseok, Baek, Seungcheol, Lee, Hyuk-Jae, Kim, John
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917809148133376
author Seo, Minseok
Nguyen, Xuan Truong
Hwang, Seok Joong
Kwon, Yongkee
Kim, Guhyun
Park, Chanwook
Kim, Ilkon
Park, Jaehan
Kim, Jeongbin
Shin, Woojae
Won, Jongsoon
Choi, Haerang
Kim, Kyuyoung
Kwon, Daehan
Jeong, Chunseok
Lee, Sangheon
Choi, Yongseok
Byun, Wooseok
Baek, Seungcheol
Lee, Hyuk-Jae
Kim, John
author_facet Seo, Minseok
Nguyen, Xuan Truong
Hwang, Seok Joong
Kwon, Yongkee
Kim, Guhyun
Park, Chanwook
Kim, Ilkon
Park, Jaehan
Kim, Jeongbin
Shin, Woojae
Won, Jongsoon
Choi, Haerang
Kim, Kyuyoung
Kwon, Daehan
Jeong, Chunseok
Lee, Sangheon
Choi, Yongseok
Byun, Wooseok
Baek, Seungcheol
Lee, Hyuk-Jae
Kim, John
contents Accelerating end-to-end inference of transformer-based large language models (LLMs) is a critical component of AI services in datacenters. However, diverse compute characteristics of end-to-end LLM inference present challenges as previously proposed accelerators only address certain operations or stages (e.g., self-attention, generation stage, etc.). To address the unique challenges of accelerating end-to-end inference, we propose IANUS -- Integrated Accelerator based on NPU-PIM Unified Memory System. IANUS is a domain-specific system architecture that combines a Neural Processing Unit (NPU) with a Processing-in-Memory (PIM) to leverage both the NPU's high computation throughput and the PIM's high effective memory bandwidth. In particular, IANUS employs a unified main memory system where the PIM memory is used both for PIM operations and for NPU's main memory. The unified main memory system ensures that memory capacity is efficiently utilized and the movement of shared data between NPU and PIM is minimized. However, it introduces new challenges since normal memory accesses and PIM computations cannot be performed simultaneously. Thus, we propose novel PIM Access Scheduling that manages normal memory accesses and PIM computations through workload mapping and scheduling across the PIM and the NPU. Our detailed simulation evaluations show that IANUS improves the performance of GPT-2 by 6.2$\times$ and 3.2$\times$, on average, compared to the NVIDIA A100 GPU and the state-of-the-art accelerator. As a proof-of-concept, we develop a prototype of IANUS with a commercial PIM, NPU, and an FPGA-based PIM controller to demonstrate the feasibility of IANUS.
format Preprint
id arxiv_https___arxiv_org_abs_2410_15008
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle IANUS: Integrated Accelerator based on NPU-PIM Unified Memory System
Seo, Minseok
Nguyen, Xuan Truong
Hwang, Seok Joong
Kwon, Yongkee
Kim, Guhyun
Park, Chanwook
Kim, Ilkon
Park, Jaehan
Kim, Jeongbin
Shin, Woojae
Won, Jongsoon
Choi, Haerang
Kim, Kyuyoung
Kwon, Daehan
Jeong, Chunseok
Lee, Sangheon
Choi, Yongseok
Byun, Wooseok
Baek, Seungcheol
Lee, Hyuk-Jae
Kim, John
Hardware Architecture
Accelerating end-to-end inference of transformer-based large language models (LLMs) is a critical component of AI services in datacenters. However, diverse compute characteristics of end-to-end LLM inference present challenges as previously proposed accelerators only address certain operations or stages (e.g., self-attention, generation stage, etc.). To address the unique challenges of accelerating end-to-end inference, we propose IANUS -- Integrated Accelerator based on NPU-PIM Unified Memory System. IANUS is a domain-specific system architecture that combines a Neural Processing Unit (NPU) with a Processing-in-Memory (PIM) to leverage both the NPU's high computation throughput and the PIM's high effective memory bandwidth. In particular, IANUS employs a unified main memory system where the PIM memory is used both for PIM operations and for NPU's main memory. The unified main memory system ensures that memory capacity is efficiently utilized and the movement of shared data between NPU and PIM is minimized. However, it introduces new challenges since normal memory accesses and PIM computations cannot be performed simultaneously. Thus, we propose novel PIM Access Scheduling that manages normal memory accesses and PIM computations through workload mapping and scheduling across the PIM and the NPU. Our detailed simulation evaluations show that IANUS improves the performance of GPT-2 by 6.2$\times$ and 3.2$\times$, on average, compared to the NVIDIA A100 GPU and the state-of-the-art accelerator. As a proof-of-concept, we develop a prototype of IANUS with a commercial PIM, NPU, and an FPGA-based PIM controller to demonstrate the feasibility of IANUS.
title IANUS: Integrated Accelerator based on NPU-PIM Unified Memory System
topic Hardware Architecture
url https://arxiv.org/abs/2410.15008