Self-Sketching Domain Specific Accelerators: Build Hardware from Software
自绘制领域特定加速器:从软件构建硬件
基本信息
- 批准号:RGPIN-2018-06795
- 负责人:
- 金额:$ 5.97万
- 依托单位:
- 依托单位国家:加拿大
- 项目类别:Discovery Grants Program - Individual
- 财政年份:2022
- 资助国家:加拿大
- 起止时间:2022-01-01 至 2023-12-31
- 项目状态:已结题
- 来源:
- 关键词:
项目摘要
While it is not always as apparent as open-source software and application innovation, computing hardware has been a key enabler for machine intelligence and AI, delivering 5000$\times$ performance improvement since 1987. Experts rank faster computing hardware as being nearly as important as neural networks for advancements in machine intelligence. Even more rapid growth in computing performance can fuel innovation in virtual reality, autonomous vehicles, and neural nets helping solve some of the grand challenges of the 21st century.Unfortunately, the hardware industry faces a stiff challenge, today while we get more transistors on a single chip, chips themselves are more expensive (\$/$mm^2$) and transistors are energy inefficient; these limit performance. The lack of a clear forecast for technology and the rise of AI has entangled companies in the messy task of creating their own custom accelerator chips to deliver the requisite performance. Unfortunately, it requires significant effort (multiple years) and money (10s of million dollars) for developing the accelerator chip and software bring-up, inaccessible to anyone else but to the largest vendors. It is unclear how accelerator development can track the rapid pace of software and application evolution. My long-term goal is, how to develop systems to make up for the lack of technology scaling. Our goal is to accelerate system architecture innovation and make it sufficiently open and inexpensive that anyone (even small vendors) can build hardware anywhere. To achieve this goal and navigate the uncertain hardware customization landscape we propose to "build hardware from software". We will be developing open-source tools and compilers, which similar to generating an CPU binary, will now generate the specification for the hardware accelerator. The key novelty of our work is the observation that hardware design should separate application domain insights from the low-level implementation details, and that the domain insights can be captured from the data types in the program. This will enable our tools to generate hardware that is reusable, shareable across algorithms, optimizable using a tool, and can be tuned by a feedback-driven process. This work will fundamentally change how systems research is done. When CPUs and GPUs dominated the hardware landscape, software has assumed hardware to be a fixed design that cannot be changed. Our work leads application experts and software developers to question the status-quo and accelerates the movement towards customization.
虽然它并不总是像开源软件和应用程序创新那样明显,但计算硬件一直是机器智能和人工智能的关键推动因素,自1987年以来提供了5000美元\倍的性能改进。对于机器智能的进步,专家们认为更快的计算硬件几乎与神经网络一样重要。计算性能的更快增长可以推动虚拟现实、自动驾驶汽车和神经网络的创新,帮助解决21世纪的一些重大挑战。不幸的是,硬件行业面临着严峻的挑战,尽管我们在一块芯片上获得了更多的晶体管,但芯片本身更昂贵(\$/$mm^2$),而且晶体管的能效很低;这些限制了性能。由于对技术缺乏明确的预测,加上人工智能的崛起,各公司陷入了创造自己的定制加速器芯片以提供必要性能的杂乱任务中。不幸的是,开发加速器芯片和软件升级需要大量的努力(多年)和资金(1000万美元),除了最大的供应商外,其他任何人都无法获得。目前尚不清楚加速器开发如何跟踪软件和应用程序发展的快速步伐。我的长期目标是,如何开发系统来弥补技术可伸缩性的不足。我们的目标是加速系统架构创新,使其足够开放和廉价,任何人(即使是小供应商)都可以在任何地方构建硬件。为了实现这一目标并驾驭不确定的硬件定制前景,我们提出了“从软件构建硬件”。我们将开发开源工具和编译器,类似于生成CPU二进制文件,现在将生成硬件加速器的规范。我们工作的主要新颖性是观察到硬件设计应该将应用程序领域洞察与低级别实现细节分开,并且可以从程序中的数据类型捕获领域洞察。这将使我们的工具能够生成可重复使用、可跨算法共享、可使用工具优化的硬件,并且可以通过反馈驱动的过程进行调整。这项工作将从根本上改变系统研究的方式。当CPU和GPU主宰硬件版图时,软件已经假定硬件是不能改变的固定设计。我们的工作导致应用程序专家和软件开发人员对现状提出质疑,并加速了向定制的运动。
项目成果
期刊论文数量(0)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
数据更新时间:{{ journalArticles.updateTime }}
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
数据更新时间:{{ journalArticles.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ monograph.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ sciAawards.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ conferencePapers.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ patent.updateTime }}
Shriraman, Arrvindh其他文献
Shriraman, Arrvindh的其他文献
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
{{ truncateString('Shriraman, Arrvindh', 18)}}的其他基金
Self-Sketching Domain Specific Accelerators: Build Hardware from Software
自绘制领域特定加速器:从软件构建硬件
- 批准号:
RGPIN-2018-06795 - 财政年份:2021
- 资助金额:
$ 5.97万 - 项目类别:
Discovery Grants Program - Individual
Self-Sketching Domain Specific Accelerators: Build Hardware from Software
自绘制领域特定加速器:从软件构建硬件
- 批准号:
RGPIN-2018-06795 - 财政年份:2020
- 资助金额:
$ 5.97万 - 项目类别:
Discovery Grants Program - Individual
Optimizing hadoop to scale to big systems and big-data
优化 hadoop 以扩展到大系统和大数据
- 批准号:
485325-2015 - 财政年份:2019
- 资助金额:
$ 5.97万 - 项目类别:
Collaborative Research and Development Grants
Self-Sketching Domain Specific Accelerators: Build Hardware from Software
自绘制领域特定加速器:从软件构建硬件
- 批准号:
RGPIN-2018-06795 - 财政年份:2019
- 资助金额:
$ 5.97万 - 项目类别:
Discovery Grants Program - Individual
Self-Sketching Domain Specific Accelerators: Build Hardware from Software
自绘制领域特定加速器:从软件构建硬件
- 批准号:
RGPIN-2018-06795 - 财政年份:2018
- 资助金额:
$ 5.97万 - 项目类别:
Discovery Grants Program - Individual
High performance computer vision on low performance hardware
低性能硬件上的高性能计算机视觉
- 批准号:
522765-2018 - 财政年份:2018
- 资助金额:
$ 5.97万 - 项目类别:
Engage Grants Program
Optimizing hadoop to scale to big systems and big-data
优化 hadoop 以扩展到大系统和大数据
- 批准号:
485325-2015 - 财政年份:2017
- 资助金额:
$ 5.97万 - 项目类别:
Collaborative Research and Development Grants
Programmable memory systems for manycore architectures
适用于众核架构的可编程内存系统
- 批准号:
402849-2011 - 财政年份:2017
- 资助金额:
$ 5.97万 - 项目类别:
Discovery Grants Program - Individual
Software framework for Smart building energy audits
智能建筑能源审计的软件框架
- 批准号:
498931-2016 - 财政年份:2016
- 资助金额:
$ 5.97万 - 项目类别:
Engage Grants Program
Optimizing hadoop to scale to big systems and big-data
优化 hadoop 以扩展到大系统和大数据
- 批准号:
485325-2015 - 财政年份:2016
- 资助金额:
$ 5.97万 - 项目类别:
Collaborative Research and Development Grants
相似海外基金
Collaborative Research: AF: Medium: Sketching for privacy and privacy for sketching
合作研究:AF:中:为隐私而素描和为素描而隐私
- 批准号:
2311649 - 财政年份:2023
- 资助金额:
$ 5.97万 - 项目类别:
Continuing Grant
Collaborative Research: HCC: Small: RUI: Drawing from Life in Extended Reality: Advancing and Teaching Cross-Reality User Interfaces for Observational 3D Sketching
合作研究:HCC:小型:RUI:从扩展现实中的生活中汲取灵感:推进和教授用于观察 3D 草图绘制的跨现实用户界面
- 批准号:
2326998 - 财政年份:2023
- 资助金额:
$ 5.97万 - 项目类别:
Standard Grant
Collaborative Research: HCC: Small: RUI: Drawing from Life in Extended Reality: Advancing and Teaching Cross-Reality User Interfaces for Observational 3D Sketching
合作研究:HCC:小型:RUI:从扩展现实中的生活中汲取灵感:推进和教授用于观察 3D 草图绘制的跨现实用户界面
- 批准号:
2326999 - 财政年份:2023
- 资助金额:
$ 5.97万 - 项目类别:
Standard Grant
Collaborative Research: AF: Medium: Sketching for privacy and privacy for sketching
合作研究:AF:中:为隐私而素描和为素描而隐私
- 批准号:
2311648 - 财政年份:2023
- 资助金额:
$ 5.97万 - 项目类别:
Continuing Grant
CAREER: Sketching for Secure Computation on Large Inputs
职业:绘制大输入安全计算草图
- 批准号:
2144798 - 财政年份:2022
- 资助金额:
$ 5.97万 - 项目类别:
Continuing Grant
Improved genomic sketching for MUMmer and metagenomics
改进了 MUMmer 和宏基因组的基因组草图
- 批准号:
10453031 - 财政年份:2022
- 资助金额:
$ 5.97万 - 项目类别:
Improved genomic sketching for MUMmer and metagenomics
改进了 MUMmer 和宏基因组的基因组草图
- 批准号:
10670162 - 财政年份:2022
- 资助金额:
$ 5.97万 - 项目类别:
Leveraging k-mer sketching statistics to enhance metagenomic methods and alignment algorithms
利用 k-mer 草图统计来增强宏基因组方法和比对算法
- 批准号:
10675449 - 财政年份:2022
- 资助金额:
$ 5.97万 - 项目类别:
CAREER: Frontiers in Matrix Sketching
职业:矩阵草图的前沿
- 批准号:
2045590 - 财政年份:2021
- 资助金额:
$ 5.97万 - 项目类别:
Continuing Grant
Collaborative Research: MLWiNS: Distributed Learning over Multi-Access Channels: From Bandlimited Coordinate Descent to Gradient Sketching
协作研究:MLWiNS:多访问通道上的分布式学习:从带限坐标下降到梯度草图
- 批准号:
2203412 - 财政年份:2021
- 资助金额:
$ 5.97万 - 项目类别:
Standard Grant