DocumentCode :
235771
Title :
Quantitative analysis for noise generated from share circuitries within DDR3 DRAM
Author :
Nam, Ilku ; Lim, Jungyoul ; Hwang, Han-jeong ; Cho, Kun ; Choi, Jang-Young
Author_Institution :
Memory Div., Samsung Electron. Co., Hwaseong, South Korea
fYear :
2014
fDate :
June 30 2014-July 4 2014
Firstpage :
83
Lastpage :
86
Abstract :
Correctable errors, almost single bit errors, can be induced from random data transition in DRAM. In the system, most of CEs are corrected by error correction code, but there is intermittent system down. Several literatures have reported that noise generated from share circuitries is regarded as a cause of soft failure. Noisy environments are one of the unavoidable factors in the high speed, high density, and low power DRAM. For the purpose of finding out noise source, we investigated states of share circuitries with random data transition. BLSA, power transistor, power line, and common plate were also researched with DDR3 DRAMs. A simple model was proposed with the quantitative analysis. Results show that the soft failure occurs when unexpected combination of noise factors happen at once, because the revealed erratic bits have similar characteristics of normal bits, except for the influence on noise.
Keywords :
DRAM chips; amplifiers; error correction codes; integrated circuit design; integrated circuit noise; low-power electronics; network analysis; power transistors; radiation hardening (electronics); BLSA; CE; DDR3 DRAM; bitline sense amplifier; correctable errors; data transition; erratic bits; error correction code; low power DRAM; noise factors; noisy environments; power line; power transistor; quantitative analysis; share circuitries; single bit errors; soft failure; Decision support systems; Failure analysis; Integrated circuits;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Physical and Failure Analysis of Integrated Circuits (IPFA), 2014 IEEE 21st International Symposium on the
Conference_Location :
Marina Bay Sands
ISSN :
1946-1542
Print_ISBN :
978-1-4799-3931-2
Type :
conf
DOI :
10.1109/IPFA.2014.6898126
Filename :
6898126
Link To Document :
بازگشت