458 0

Analysis of Radiation Effect to Row Hammer Fault in SDRAMs

Analysis of Radiation Effect to Row Hammer Fault in SDRAMs
Other Titles
방사선 데미지 영향에 의한 SDRAM 상에서의 로우 해머링 고장 분석 연구
Alternative Author(s)
Chulseung Lim
Issue Date
As technology scaling, Dynamic Random Access Memory (DRAM) has been rapidly developed to enhance operating speed and bandwidth for various applications. Because Synchronous DRAM (SDRAM) has a relatively simple structure, it enables high integration density. A DRAM cell suffers from intrinsic leakage current from the capacitive storage element. The leakage current is aggravated further by technology scaling. Therefore, cells need to be regularly refreshed to sustain their stored data. As technology scaling, the size of the storage capacitor has been scaled down, and the Joint Electron Device Engineering Council (JEDEC) standard recommends a 64 ms of tRET. This dissertation proposes the row hammering test to evaluate displacement damage (DD) in SDRAM components. Irradiated SDRAM devices could have multiple current leakage paths, owing to DD effects. The degree of leakage from cells with DD can be differentiated from pre-irradiation row hammering test results. Proton-based SER tests were performed with DDR3 and DDR4 SDRAM components made using 3x-nm, and 2x-nm technologies. The experimental results showed that failed cells caused by the row hammering test were more than five times higher in the irradiated sample compared to the non-irradiated sample, and failed cells by row hammer were not detectable within 64-ms retention time using the traditional retention test method. In the worst case, the number of hammerings required to cause a row hammering-induced failure in the irradiated sample reduced by 36 times compared to that in the non-irradiated sample even after the irradiated sample was annealed at 150°C so as to have no retention errors within the maximum retention time of 64 ms. In DDR4 SDRAMs, experiment results showed that after proton irradiation, the number of bit errors caused by the row hammering test increased about 41% and 66% in technologies 2x-nm and 2y-nm, respectively. With 2y-nm technology, bit errors started to appear from 5K Number of Hammering (NHMR)—the equivalent of 500 μs retention time. For 2y-nm components, more than 90% of failed words had multiple failed cells, which could not be corrected by Single Error Correction and Double Error Detection (SEC-DED) codes. In this dissertation suggests that the row hammering test could be one of efficient test method in active way compared with the retention test which is for passive test method. The row hammering test could detect the cells which can be possibly failed in early stage compare than the retention test.
Appears in Collections:
Files in This Item:
There are no files associated with this item.
RIS (EndNote)
XLS (Excel)


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.