Close
 

Session

Research Manuscript: Resist Faults and Make Memories Smarter
Session Chairs
Arman Roohi
University of Nebraska
Shaahin Angizi
New Jersey Institute of Technology
Event TypeResearch Manuscript
Keywords
In-memory and Near-memory Computing
Topics
Design
TimeThursday, July 14th10:30am - 12:00pm PDT
Location3002, Level 3
DescriptionResistive RAM based In-Memory Computing offers a promising path to break the von-Neumann bottleneck. This session pushes the frontiers in this novel computing paradigm by exploring algorithmic techniques to address fundamental challenges related to noise and fault tolerance. Application-hardware co-design demonstrates the potential of such In-Memory computing platforms for graph based vector search and string processing applications.
Presentations
10:30am - 10:52am PDTStatistical Computing Framework and Demonstration for In-memory Computing Systems
Authors
Bonan Zhang
Peter Deaville
Naveen Verma
10:52am - 11:15am PDTWrite or Not: Programming Scheme Optimization for RRAM-based Neuromorphic Computing
Authors
Ziqi Meng
Yanan Sun
Weikang Qian
11:15am - 11:37am PDTReSMA: Accelerating Approximate String Matching Using ReRAM-based Content Addressable Memory
Authors
Huize Li
Hai Jin
Long Zheng
Yu Huang
Xiaofei Liao
Zhuohui Duan
Dan Chen
Chuangyi Gui
11:37am - 12:00pm PDTVStore: In-Storage Graph Based Vector Search Accelerator
Authors
Shengwen Liang
Ziming Yuan
Ying Wang
Cheng Liu
Huawei Li
Xiaowei Li