Is the concern for soft-error overblown?

Sandip Kundu; Rajesh Galivanche; Vijaykrishnan Narayanan; Rajesh Raina; Pia Sanda; Kaushik Roy

doi:10.1109/TEST.2005.1584100

Is the concern for soft-error overblown?

Sandip Kundu, Rajesh Galivanche, Vijaykrishnan Narayanan, Rajesh Raina, Pia Sanda, Kaushik Roy

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Scopus citations

Abstract

Soft-Error caused by ionizing radiation has emerged as a cause for great concern. It is argued that with scaling of transistor feature size, soft-error rate will increase by orders of magnitude and special solutions will be needed to deal with this problem. However, anecdotal evidence suggests soft-error is not the leading cause of transient faults today; the power supply droop, unforeseen workloads, thermal hot spots, active voltage, frequency and temperature regulation related issues cause problems with far greater incidence. While a problem is a problem, is engineering focus on soft-error the right priority? Is soft-error likely to dominate transient faults in future? Should this become a central focus during design process?

Original language	English (US)
Title of host publication	IEEE International Test Conference, Proceedings, ITC 2005
Pages	1268
Number of pages	1
DOIs	https://doi.org/10.1109/TEST.2005.1584100
State	Published - 2005
Event	IEEE International Test Conference, ITC 2005 - Austin, TX, United States Duration: Nov 8 2005 → Nov 10 2005

Publication series

Name	Proceedings - International Test Conference
Volume	2005
ISSN (Print)	1089-3539

Other

Other	IEEE International Test Conference, ITC 2005
Country/Territory	United States
City	Austin, TX
Period	11/8/05 → 11/10/05

All Science Journal Classification (ASJC) codes

Electrical and Electronic Engineering
Applied Mathematics

Access to Document

10.1109/TEST.2005.1584100

Cite this

@inproceedings{6b6ff3b86ee7477a86b81676bb0d7e33,

title = "Is the concern for soft-error overblown?",

abstract = "Soft-Error caused by ionizing radiation has emerged as a cause for great concern. It is argued that with scaling of transistor feature size, soft-error rate will increase by orders of magnitude and special solutions will be needed to deal with this problem. However, anecdotal evidence suggests soft-error is not the leading cause of transient faults today; the power supply droop, unforeseen workloads, thermal hot spots, active voltage, frequency and temperature regulation related issues cause problems with far greater incidence. While a problem is a problem, is engineering focus on soft-error the right priority? Is soft-error likely to dominate transient faults in future? Should this become a central focus during design process?",

author = "Sandip Kundu and Rajesh Galivanche and Vijaykrishnan Narayanan and Rajesh Raina and Pia Sanda and Kaushik Roy",

year = "2005",

doi = "10.1109/TEST.2005.1584100",

language = "English (US)",

isbn = "0780390393",

series = "Proceedings - International Test Conference",

pages = "1268",

booktitle = "IEEE International Test Conference, Proceedings, ITC 2005",

note = "IEEE International Test Conference, ITC 2005 ; Conference date: 08-11-2005 Through 10-11-2005",

}

Kundu, S, Galivanche, R, Narayanan, V, Raina, R, Sanda, P & Roy, K 2005, Is the concern for soft-error overblown? in IEEE International Test Conference, Proceedings, ITC 2005., 1584100, Proceedings - International Test Conference, vol. 2005, pp. 1268, IEEE International Test Conference, ITC 2005, Austin, TX, United States, 11/8/05. https://doi.org/10.1109/TEST.2005.1584100