Abstract:The erratic bits (EB) phenomenon in nonvolatile memory devices (NVMs) has been evidenced in several technologies as a main reliability detractor. Usually, this issue is handled by repair strategies, which spans from static redundancy to dynamic correction codes. This evidences a tradeoff in a reliability/ performance domain that is due to the limitation in the repair resources amount and correction strength. In this paper, we expose this tradeoff in different NVM technologies such as embedded NOR Flash and pha… Show more
“…However, from the pure reliability viewpoint, the erase operation has always been an issue for every NOR Flash technology so far [37]. In the last decades, special emphasis has been put on the reduction of the standard erase failure mechanisms by implementing either proper correction algorithms or failure screening techniques aiming at a memory fail rate reduction below 1 ppm [38]. Such activity involves: complex time-consuming recovery algorithms that burden on the erase latency, Error Correction Codes (ECC) design for fast error detection and correction to limit the impact on the read latency, and redundancy-based solutions that increase the costs of the final product where the NVM is embedded.…”
Section: B Nvms Operating In Harsh Environmentsmentioning
The paper critically analyzes the trends and limits of integrated technologies for sensing, computing and data storage when devices and systems, originally developed for consumer electronics, are used for self-driving cars. Some hints, supported by theoretical analysis and experimental measures, are provided to overcome the issues of inertial sensors, micro-mirrors for Lidar scanners, car data/program memories and computing platforms.
“…However, from the pure reliability viewpoint, the erase operation has always been an issue for every NOR Flash technology so far [37]. In the last decades, special emphasis has been put on the reduction of the standard erase failure mechanisms by implementing either proper correction algorithms or failure screening techniques aiming at a memory fail rate reduction below 1 ppm [38]. Such activity involves: complex time-consuming recovery algorithms that burden on the erase latency, Error Correction Codes (ECC) design for fast error detection and correction to limit the impact on the read latency, and redundancy-based solutions that increase the costs of the final product where the NVM is embedded.…”
Section: B Nvms Operating In Harsh Environmentsmentioning
The paper critically analyzes the trends and limits of integrated technologies for sensing, computing and data storage when devices and systems, originally developed for consumer electronics, are used for self-driving cars. Some hints, supported by theoretical analysis and experimental measures, are provided to overcome the issues of inertial sensors, micro-mirrors for Lidar scanners, car data/program memories and computing platforms.
“…Among them, the Random Telegraph Noise (RTN) related to filling/empting of tunnel oxide traps affects the V T distributions stability few microseconds after the application of the programming pulse, creating distribution tails below the target verification level [52], [53], [54], [55]. Additionally, positive trapped charge in the tunnel oxide during cycling results in a modified FN tunnel dynamics that may trigger erratic effects [33], [56], [57], [58]. These sporadic mechanisms, that may potentially affect any cell in the array, have a random and transient nature; they can occur during any programming pulse and they may produce threshold shifts larger than expected, with the risk of programming some cells with a threshold voltage larger than the desired one.…”
Solid-state drives (SSDs) faced an astonishing development in the last few years, becoming the cornerstone to new paradigms and markets of the information technology, such as cloud computing and big data centers. So far, the SSD design approach has focused on the optimization of the Flash translation layer, the firmware devoted to fulfill the compatibility with traditional hard-disk drives. For hyperscaled SSDs this strategy is no longer valid since their performance and reliability are strictly linked to that of the NAND Flash memories that constitute the storage medium, in particular when the multilevel cell paradigm is considered. For this reason, the design flow must follow a bottom-up approach that, starting from an accurate knowledge of the time and use dependent reliability of the NAND Flash memories, selects the most appropriate error correction strategy to extend the SSD lifetime while reducing its performance degradation. Then, the design flow moves to that of the SSD controller and of the interface toward the host where the application is running. This paper will thoroughly discuss this bottom-up approach, and finally, it will show how it is possible to leverage new approaches, such as the software-defined storage system that, by exploiting a hardware/software codesign of the SSD controller architecture and of the host application, will be able to revolutionize the traditional computer/memory interaction
“…An application of the proposed compact model turns particularly useful in the evaluation of the erase operation reliability, therefore allowing extended considerations on the best approaches to follow (i.e., soft-programming algorithms or error correction codes [4]) to increase this metric. The case study concerns the evaluation of the bitline current I read when overerased cells share the same bitline of a programmed cell to be read, as shown in Fig.…”
The simulation of the erratic bits phenomenon in Flash memory arrays for reliability projections has been a matter of study in the last decade from many standpoints. However, the majority of the developed simulation framework lacked both a direct link with the physics underlying the phenomenon and an easy integration with circuit simulators for fast analysis. In this paper, we have developed a compact model for erratic events starting from the PSP-model description of a Flash cell using Verilog-A. The model has been focused on the reproduction of the overerase phenomenon in a 90-nm NOR Flash array. Its accurate and fast simulation capabilities allowed the evaluation of the array reliability against the erratic erase operation
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.