LUTIC:一种基于ram的电源故障弹性内存计算架构

2023 26th International Symposium on Design and Diagnostics of Electronic Circuits and Systems (DDECS) Pub Date : 2023-05-03 DOI:10.1109/DDECS57882.2023.10139576

K. Akhunov, K. Yıldırım

{"title":"LUTIC:一种基于ram的电源故障弹性内存计算架构","authors":"K. Akhunov, K. Yıldırım","doi":"10.1109/DDECS57882.2023.10139576","DOIUrl":null,"url":null,"abstract":"Processing In-Memory (PIM) based on emerging non-volatile memory technologies can accelerate machine learning tasks even for batteryless devices operating intermittently. However, existing PIM solutions for intermittent systems offer limited parallelism and do not support a high degree of programmability. This paper presents LUTIC —a novel architecture that can accelerate a broader class of intermittent data-intense operations. Our results demonstrate that, compared to existing commercial low-energy accelerators and PIM solutions for intermittent computing, LUTIC improves performance and energy efficiency with better parallelism support.","PeriodicalId":220690,"journal":{"name":"2023 26th International Symposium on Design and Diagnostics of Electronic Circuits and Systems (DDECS)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"LUTIC: A CRAM-based Architecture for Power Failure Resilient In-Memory Computing\",\"authors\":\"K. Akhunov, K. Yıldırım\",\"doi\":\"10.1109/DDECS57882.2023.10139576\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Processing In-Memory (PIM) based on emerging non-volatile memory technologies can accelerate machine learning tasks even for batteryless devices operating intermittently. However, existing PIM solutions for intermittent systems offer limited parallelism and do not support a high degree of programmability. This paper presents LUTIC —a novel architecture that can accelerate a broader class of intermittent data-intense operations. Our results demonstrate that, compared to existing commercial low-energy accelerators and PIM solutions for intermittent computing, LUTIC improves performance and energy efficiency with better parallelism support.\",\"PeriodicalId\":220690,\"journal\":{\"name\":\"2023 26th International Symposium on Design and Diagnostics of Electronic Circuits and Systems (DDECS)\",\"volume\":\"50 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-05-03\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2023 26th International Symposium on Design and Diagnostics of Electronic Circuits and Systems (DDECS)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/DDECS57882.2023.10139576\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 26th International Symposium on Design and Diagnostics of Electronic Circuits and Systems (DDECS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/DDECS57882.2023.10139576","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

摘要

基于新兴非易失性存储技术的内存处理(PIM)可以加速机器学习任务，即使是间歇性运行的无电池设备。然而，现有的间歇性系统的PIM解决方案提供了有限的并行性，并且不支持高度的可编程性。本文提出了LUTIC——一种新的架构，可以加速更广泛的间歇性数据密集型操作。我们的研究结果表明，与现有的用于间歇性计算的商用低能量加速器和PIM解决方案相比，LUTIC通过更好的并行支持提高了性能和能源效率。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

LUTIC: A CRAM-based Architecture for Power Failure Resilient In-Memory Computing

Processing In-Memory (PIM) based on emerging non-volatile memory technologies can accelerate machine learning tasks even for batteryless devices operating intermittently. However, existing PIM solutions for intermittent systems offer limited parallelism and do not support a high degree of programmability. This paper presents LUTIC —a novel architecture that can accelerate a broader class of intermittent data-intense operations. Our results demonstrate that, compared to existing commercial low-energy accelerators and PIM solutions for intermittent computing, LUTIC improves performance and energy efficiency with better parallelism support.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

2023 26th International Symposium on Design and Diagnostics of Electronic Circuits and Systems (DDECS)

自引率

0.00%

发文量