Search NASA⌕ Search

SEARCH · Search NASA

Results for “flash memory”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2

Data Movement Dominates: Final Report

Over the past three years in this project, what we have observed is that the primary reason for data movement in large-scale systems is that the per-node capacity is not large enough—i.e., one of the solutions to the data-movement problem (certainly not the only solution that is required, but a significant one nonetheless) is to increase per-node capacity so that inter-node traffic is reduced. This unfortunately is not as simple as it sounds. Today’s main memory systems for datacenters, enterprise computing systems, and supercomputers, fail to provide high per-socket capacity [Dirik & Jacob 2009; Cooper-Balis et al. 2012], except at extremely high price points (factors of 10–100x the cost/bit of consumer main-memory systems) [Stokes 2008]. The reason is that our choice of technology for today’s main memory systems—i.e., DRAM, which we have used as a main-memory technology since the 1970s [Jacob et al. 2007]—can no longer keep up with our needs for density and price per bit. Main memory systems have always been built from the cheapest, densest, lowest-power memory technology available, and DRAM is no longer the cheapest, the densest, nor the lowest-power storage technology out there. It is now time for DRAM to go the way that SRAM went: move out of the way for a cheaper, slower, denser storage technology, and become a cache instead. This inflection point has happened before, in the context of SRAM yielding to DRAM. There was once a time that SRAM was the storage technology of choice for all main memories [Tomasulo 1967; Thornton 1970; Kidder 1981]. However, once DRAM hit volume production in the 1970s and 80s, it supplanted SRAM as a main memory technology because it was cheaper, and it was denser. It also happened to be lower power, but that was not the primary consideration of the day. At the time, it was recognized that DRAM was much slower than SRAM, but it was only at the supercomputer level (For instance the Cray X-MP in the 1980s and its follow-on, the Cray Y-MP, in the 1990s) that could one afford to build ever- larger main memories out of SRAM—the reasoning for moving to DRAM was that an appropriately designed memory hierarchy, built of DRAM as main memory and SRAM as a cache, would approach the performance of SRAM, at the price-per-bit of DRAM [Mashey 1999]. Today it is quite clear that, were one to build an entire multi-gigabyte main memory out of SRAM instead of DRAM, one could improve the performance of almost any computer system by up to an order of magnitude—but this option is not even considered, because to build that system would be prohibitively expensive. It is now time to revisit the same design choice in the context of modern technologies and modern systems. For reasons both technical and economic, we can no longer afford to build ever-larger main memory systems out of DRAM. Flash memory, on the other hand, is significantly cheaper and denser than DRAM and therefore should take its place. While it is true that flash is significantly slower than DRAM, one can afford to build much larger main memories out of flash than out of DRAM, and we show that an appropriately designed memory hierarchy, built of flash as main memory and DRAM as a cache, will approach the performance of DRAM, at the price-per-bit of flash. In our studies as part of this project, we have investigated Non-Volatile Main Memory (NVMM), a new main-memory architecture for large-scale computing systems, one that is specifically designed to address the weaknesses described previously. In particular, it provides the following features: non-volatility: The bulk of the storage is comprised of NAND flash, and in this organization DRAM is used only as a cache, not as main memory. Furthermore, the flash is journaled, which means that operations such as checkpoint/restore are already built into the system. 1+ terabytes of storage per socket: SSDs and DRAM DIMMs have roughly the same form factor (several square inches of PCB surface area), and terabyte SSDs are now commonplace. performance approaching that of DRAM: DRAM is used as a cache to the flash system. price-per-bit approaching that of NAND: Flash is currently well under $0.50 per gigabyte; DDR3 SDRAM is currently just over $10 per gigabyte [Newegg 2014]. Even today, one can build an easily affordable main memory system with a terabyte or more of NAND storage per CPU socket (which would be extremely expensive were one to use DRAM), and our cycle- accurate, full-system experiments show that this can be done at a performance point that lies within a factor of two of DRAM.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

FPGA remote update for nuclear environments

The Instituto de Plasmas e Fusao Nuclear (IPFN) has developed dedicated re-configurable modules based on field programmable gate array (FPGA) devices for several nuclear fusion machines worldwide. Moreover, new Advanced Telecommunication Computing Architecture (ATCA) based modules developed by IPFN are already included in the ITER catalogue. One of the requirements for re-configurable modules operating in future nuclear environments including ITER is the remote update capability. Accordingly, this work presents an alternative method for FPGA remote programing to be implemented in new ATCA based re-configurable modules. FPGAs are volatile devices and their programming code is usually stored in dedicated flash memories for properly configuration during module power-on. The presented method is capable to store new FPGA codes in Serial Peripheral Interface (SPI) flash memories using the PCIexpress (PCIe) network established on the ATCA back-plane, linking data acquisition endpoints and the data switch blades. The method is based on the Xilinx Quick Boot application note, adapted to PCIe protocol and ATCA based modules. (authors)

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

ASA-FTL: An adaptive separation aware flash translation layer for solid state drives

Here, the flash-memory based Solid State Drive (SSD) presents a promising storage solution for increasingly critical data-intensive applications due to its low latency (high throughput), high bandwidth, and low power consumption. Within an SSD, its Flash Translation Layer (FTL) is responsible for exposing the SSD’s flash memory storage to the computer system as a simple block device. The FTL design is one of the dominant factors determining an SSD’s lifespan and performance. To reduce the garbage collection overhead and deliver better performance, we propose a new, low-cost, adaptive separation-aware flash translation layer (ASA-FTL) that combines sampling, data clustering and selective caching of recency information to accurately identify and separate hot/cold data while incurring minimal overhead. We use sampling for light-weight identification of separation criteria, and our dedicated selective caching mechanism is designed to save the limited RAM resource in contemporary SSDs. Using simulations of ASA-FTL with both real-world and synthetic workloads, we have shown that our proposed approach reduces the garbage collection overhead by up to 28% and the overall response time by 15% compared to one of the most advanced existing FTLs. We find that the data clustering using a small sample size provides significant performance benefit while only incurring a very small computation and memory cost. In addition, our evaluation shows that ASA-FTL is able to adapt to the changes in the access pattern of workloads, which is a major advantage comparing to existing fixed data separation methods.

97 MATHEMATICS AND COMPUTING↗

High Performance Data Transfer for Distributed Data Intensive Sciences

We report on the development of ZX software providing high performance data transfer and encryption. The design scales in: computation power, network interfaces, and IOPS while carefully balancing the available resources. Two U.S. patent-pending algorithms help tackle data sets containing lots of small files and very large files, and provide insensitivity to network latency. It has a cluster-oriented architecture, using peer-to-peer technologies to ease deployment, operation, usage, and resource discovery. Its unique optimizations enable effective use of flash memory. Using a pair of existing data transfer nodes at SLAC and NERSC, we compared its performance to that of bbcp and GridFTP and determined that they were comparable. With a proof of concept created using two four-node clusters with multiple distributed multi-core CPUs, network interfaces and flash memory, we achieved 155Gbps memory-to-memory over a 2x100Gbps link aggregated channel and 70Gbps file-to-file with encryption over a 5000 mile 100Gbps link.

97 MATHEMATICS AND COMPUTING↗

Resistive memory for radiation resistant non-volatile memory (NSTRF Final Report)

As space programs increase in number and scope, there is an increasing need for radiation-hardened electronic devices and circuits. In particular, missions to high-radiation environments, such as Europa, would greatly benefit from improved radiation hardness in electronic devices. In pursuit of this goal, resistive memory (RRAM) devices were fabricated at SUNY Polytechnic Institute and evaluated for radiation hardness. Our objectives were to produce RRAM devices resistant to high levels of radiation damage and to demonstrate that these devices would improve mission lifetime in high-radiation environments. Furthermore, the underlying mechanisms of radiation were investigated to provide recommendations for radiation-hardening RRAM devices, which could be applied to any candidate RRAM devices being considered for space applications. Devices were fabricated using several fabrication approaches, including patterning by shadow mask, photolithography-based etching, and photolithography-based liftoff. In each of these cases, total ionizing dose (TID) effects and displacement damage dose (DDD) effects were measured. TID effects from exposure to a 60Co gamma source were not observed to cause changes in device resistance or switching parameters in any experiments, with each device tested to at least 20 Mrad(Si). DDD was measured as radiation-generated oxygen vacancies per cm 3 since oxygen vacancies are generally considered to be the active species involved in switching these devices. The lowest DDD level that caused a device to change resistance state was 10 21 vacancies per cm 3 , and most devices failed at 1022 vacancies per cm 3 . This is an extremely high DDD level, even for RRAM devices, which have been reported to fail in the range of 10 17 -10 20 vacancies per cm 3 . For comparison, an example flash memory device failed at 10 15 vacancies per cm 3 . Vendor-fabricated devices with a similar composition to our own were also tested against TID and DDD. The vendor-fabricated devices did not exhibit changes due to TID, up to the tested level of 30 Mrad(Si). Meanwhile, vendor devices exhibited resistance state changes at 10 21 vacancies per cm 3 , similar to our own devices. These results indicate that Ta0x-based RRAM devices may be particularly resilient to both TID and DDD effects. The very high tolerance to radiation effects is most likely due to the high intrinsic concentration of oxygen vacancies within our devices. Based on X-ray photoelectron spectroscopy (XPS) measurements, there are approximately 10 22 oxygen vacancies per cm 3 in our devices as deposited. Most devices failed when the radiation-induced vacancies reached this level, indicating suggesting that a high intrinsic vacancy concentration protects against lower levels of displacement damage. High vacancy concentration likely also protects against TID by facilitating leakage of trapped charge out of the oxide. The use of a thin switching oxide (25 nm Ta0 x , for our devices) is also expected to improve radiation hardness, as there is less room for charge trapping. Therefore, those wishing to produce very radiation-tolerant RRAM devices can probably achieve this by using a thin oxide that contains a high intrinsic concentration of oxygen vacancies. Our devices appear to be very tolerant of radiation effects, and would greatly increase the expected lifetime of a mission to Europa or another high-radiation target compared to flash memory devices. The similar radiation performance of vendor-fabricated devices is promising for adoption of RRAM devices as radiation-hardened memory devices for use in space. With continued commercial development of these devices, RRAM devices are strong candidates for next-generation memories that are inherently rad-hard.

36 MATERIALS SCIENCE↗

Exploiting Internal Parallelism for Address Translation in Solid-State Drives

Solid-state Drives (SSDs) have changed the landscape of storage systems and present a promising storage solution for data-intensive applications due to their low latency, high bandwidth, and low power consumption compared to traditional hard disk drives. SSDs achieve these desirable characteristics using internal parallelism—parallel access to multiple internal flash memory chips—and a Flash Translation Layer (FTL) that determines where data are stored on those chips so that they do not wear out prematurely. However, current state-of-the-art cache-based FTLs like the Demand-based Flash Translation Layer (DFTL) do not allow IO schedulers to take full advantage of internal parallelism, because they impose a tight coupling between the logical-to-physical address translation and the data access. In this study to address this limitation, we introduce a new FTL design called Parallel-DFTL that works with the DFTL to decouple address translation operations from data accesses. Parallel-DFTL separates address translation and data access operations into different queues, allowing the SSD to use concurrent flash accesses for both types of operations. We also present a Parallel-LRU cache replacement algorithm to improve the concurrency of address translation operations. To compare Parallel-DFTL against existing FTL approaches, we present a Parallel-DFTL performance model and compare its predictions against those for DFTL and an ideal page-mapping approach. We also implemented the Parallel-DFTL approach in an SSD simulator using real device parameters, and used trace-driven simulation to evaluate Parallel-DFTL’s efficacy. Our evaluation results show that Parallel-DFTL improved the overall performance by up to 32% for the real IO workloads we tested, and by up to two orders of magnitude with synthetic test workloads. Finally, we also found that Parallel-DFTL is able to achieve reasonable performance with a very small cache size and that it provides the best benefit for those workloads with large request size or with high write ratio.

97 MATHEMATICS AND COMPUTING↗

Non-volatile memory for checkpoint storage

A system, method and computer program product for supporting system initiated checkpoints in high performance parallel computing systems and storing of checkpoint data to a non-volatile memory storage device. The system and method generates selective control signals to perform checkpointing of system related data in presence of messaging activity associated with a user application running at the node. The checkpointing is initiated by the system such that checkpoint data of a plurality of network nodes may be obtained even in the presence of user applications running on highly parallel computers that include ongoing user messaging activity. In one embodiment, the non-volatile memory is a pluggable flash memory card.

97 MATHEMATICS AND COMPUTING↗

Total-Ionizing-Dose Effects on Long-term Data Retention Characteristics of Commercial 3-D NAND Memories

This article evaluates the data retention characteristics of irradiated multi-level cell (MLC) 3-D NAND flash memories. We irradiate the memory chips by a Co-60 gamma-ray source for up to 50 krad(Si) and then write a random data pattern on the irradiated chip to find its retention characteristics. The experimental results show that the data retention property of the irradiated chips is significantly degraded compared to the un-irradiated ones. We evaluate two independent strategies to improve the data retention characteristics of the irradiated chips. Furthermore, the first method involves high-temperature annealing of the irradiated chip while the second method suggests pre-programming the memory modules before deploying them in radiation-prone environments.

3-D NAND flash↗

A Semi-Preemptive Garbage Collector for Solid State Drives

NAND flash memory is a preferred storage media for various platforms ranging from embedded systems to enterprise-scale systems. Flash devices do not have any mechanical moving parts and provide low-latency access. They also require less power compared to rotating media. Unlike hard disks, flash devices use out-of-update operations and they require a garbage collection (GC) process to reclaim invalid pages to create free blocks. This GC process is a major cause of performance degradation when running concurrently with other I/O operations as internal bandwidth is consumed to reclaim these invalid pages. The invocation of the GC process is generally governed by a low watermark on free blocks and other internal device metrics that different workloads meet at different intervals. This results in I/O performance that is highly dependent on workload characteristics. In this paper, we examine the GC process and propose a semi-preemptive GC scheme that can preempt on-going GC processing and service pending I/O requests in the queue. Moreover, we further enhance flash performance by pipelining internal GC operations and merge them with pending I/O requests whenever possible. Our experimental evaluation of this semi-preemptive GC sheme with realistic workloads demonstrate both improved performance and reduced performance variability. Write-dominant workloads show up to a 66.56% improvement in average response time with a 83.30% reduced variance in response time compared to the non-preemptive GC scheme.

99 GENERAL AND MISCELLANEOUS↗

Low-temperature etching of silicon oxide and silicon nitride with hydrogen fluoride

Etching of high aspect ratio features into alternating SiO 2 and SiN layers is an enabling technology for the manufacturing of 3D NAND flash memories. In this paper, we study a low-temperature or cryo plasma etch process, which utilizes HF gas together with other gas additives. Compared with a low-temperature process that uses separate fluorine and hydrogen gases, the etching rate of the SiO 2 /SiN stack doubles. Both materials etch faster with this so-called second generation cryo etch process. Pure HF plasma enhances the SiN etching rate, while SiO 2 requires an additional fluorine source such as PF 3 to etch meaningfully. Further, the insertion of H 2 O plasma steps into the second generation cryo etch process boosts the SiN etching rate by a factor of 2.4, while SiO 2 etches only 1.3 times faster. We observe a rate enhancing effect of H 2 O coadsorption in thermal etching experiments of SiN with HF. Ammonium fluorosilicate (AFS) plays a salient role in etching of SiN with HF with and without plasma. AFS appears weakened in the presence of H 2 O. Density functional theory calculations confirm the reduction of the bonding energy when NH 4 F in AFS is replaced by H 2 O.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Local structure effects of carbon-doping on the phase change material Ge 2 Sb 2 Te 5

Ge 2 Sb 2 Te 5 is used in phase change memory, a nonvolatile memory technology, due to its phase change properties. The primary advantage of phase change memory over the state-of-the-art (flash memory) is its simple and small device geometry, which allows for denser nodes and lower power consumption. In phase change memory, resistive heating induces fast switching between the high resistance amorphous and low resistance crystalline phases, corresponding to storage of low and high digital states, respectively. However, the instability of the amorphous phase of Ge 2 Sb 2 Te 5 presents issues with processing and long-term data storage; such issues can be resolved by C doping, which stabilizes the amorphous phase and raises the crystallization temperature. To better understand the local structural effects of C doping on Ge 2 Sb 2 Te 5 , in situ Ge K-edge X-ray absorption spectroscopy measurements were taken during heating of films with various C doping concentrations. Here, the range of structural transformation temperatures derived from X-ray absorption near-edge structure analysis across the C doping series proved narrower than crystallization temperatures reported in similar in situ X-ray diffraction experiments, which may reflect changes in local structure that precede long-range ordering during crystallization. In addition, rigorous extended X-ray absorption fine structure fitting across and between temperature series revealed effects of C doping on the rigidity of Ge–Te bonds at low (2 at% and 4 at%) C concentrations.

Langhout, John D.↗

On the origin of resistive switching volatility in Ni/TiO{sub 2}/Ni stacks

Resistive switching and resistive random access memories have attracted huge interest for next generation nonvolatile memory applications, also thought to be able to overcome flash memories limitations when arranged in crossbar arrays. A cornerstone of their potential success is that the toggling between two distinct resistance states, usually a High Resistive State (HRS) and a Low Resistive State (LRS), is an intrinsic non-volatile phenomenon with the two states being thermodynamically stable. TiO{sub 2} is one of the most common materials known to support non-volatile RS. In this paper, we report a volatile resistive switching in a titanium dioxide thin film sandwiched by two nickel electrodes. The aim of this work is to understand the underlying physical mechanism that triggers the volatile effect, which is ascribed to the presence of a NiO layer at the bottom interface. The NiO layer alters the equilibrium between electric field driven filament formation and thermal enhanced ion diffusion, resulting in the volatile behaviour. Although the volatility is not ideal for non-volatile memory applications, it shows merit for access devices in crossbar arrays due to its high LRS/HRS ratio, which are also briefly discussed.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Checkpoint-Restart in User Space

CRUISE implements a user-space file system that stores data in main memory and transparently spills over to other storage, like local flash memory or the parallel file system, as needed. CRUISE also exposes file contents fo remote direct memory access, allowing external tools to copy files to the parallel file system in the background with reduced CPU interruption.

Source record↗

Checkpoint-Restart in User Space

CRUISE implements a user-space file system that stores data in main memory and transparently spills over to other storage, like local flash memory or the parallel file system, as needed. CRUISE also exposes file contents fo remote direct memory access, allowing external tools to copy files to the parallel file system in the background with reduced CPU interruption.

Source record↗

A Layered Solution for Supercomputing Storage

To solve the supercomputing challenge of memory keeping up with processing speed, a team at Los Alamos National Laboratory developed two innovative memory management and storage technologies. Burst buffers peel off data onto flash memory to support the checkpoint/restart paradigm of large simulations. MarFS adds a thin software layer enabling a new tier for campaign storage—based on inexpensive, failure-prone disk drives—between disk drives and tape archives.

97 MATHEMATICS AND COMPUTING↗

Impact of metal nano layer thickness on tunneling oxide and memory performance of core-shell iridium-oxide nanocrystals

The impact of iridium-oxide (IrO{sub x}) nano layer thickness on the tunneling oxide and memory performance of IrO{sub x} metal nanocrystals in an n-Si/SiO{sub 2}/Al{sub 2}O{sub 3}/IrO{sub x}/Al{sub 2}O{sub 3}/IrO{sub x} structure has been investigated. A thinner (1.5 nm) IrO{sub x} nano layer has shown better memory performance than that of a thicker one (2.5 nm). Core-shell IrO{sub x} nanocrystals with a small average diameter of 2.4 nm and a high density of {approx}2 x 10{sup 12}/cm{sup 2} have been observed by scanning transmission electron microscopy. The IrO{sub x} nanocrystals are confirmed by x-ray photoelectron spectroscopy. A large memory window of 3.0 V at a sweeping gate voltage of {+-}5 V and 7.2 V at a sweeping gate voltage of {+-} 8 V has been observed for the 1.5 nm-thick IrO{sub x} nano layer memory capacitors with a small equivalent oxide thickness of 8 nm. The electrons and holes are trapped in the core and annular regions of the IrO{sub x} nanocrystals, respectively, which is explained by Gibbs free energy. High electron and hole-trapping densities are found to be 1.5 x 10{sup 13}/cm{sup 2} and 2 x 10{sup 13}/cm{sup 2}, respectively, due to the small size and high-density of IrO{sub x} nanocrystals. Excellent program/erase endurance of >10{sup 6} cycles and good retention of 10{sup 4} s with a good memory window of >1.2 V under a small operation voltage of {+-} 5 V are obtained. A large memory size of >10 Tbit/sq. in. can be designed by using the IrO{sub x} nanocrystals. This study is not only important for the IrO{sub x} nanocrystal charge-trapping memory investigation but it will also help to design future metal nanocrystal flash memory.

36 MATERIALS SCIENCE↗

Pressure tunes electrical resistivity by four orders of magnitude in amorphous Ge[subscript 2]Sb[subscript 2]Te[subscript 5] phase-change memory alloy

Ge-Sb-Te-based phase-change memory is one of the most promising candidates to succeed the current flash memories. The application of phase-change materials for data storage and memory devices takes advantage of the fast phase transition (on the order of nanoseconds) and the large property contrasts (e.g., several orders of magnitude difference in electrical resistivity) between the amorphous and the crystalline states. Despite the importance of Ge-Sb-Te alloys and the intense research they have received, the possible phases in the temperature-pressure diagram, as well as the corresponding structure-property correlations, remain to be systematically explored. In this study, by subjecting the amorphous Ge{sub 2}Sb{sub 2}Te{sub 5} (a-GST) to hydrostatic-like pressure (P), the thermodynamic variable alternative to temperature, we are able to tune its electrical resistivity by several orders of magnitude, similar to the resistivity contrast corresponding to the usually investigated amorphous-to-crystalline (a-GST to rock-salt GST) transition used in current phase-change memories. In particular, the electrical resistivity drops precipitously in the P = 0 to 8 GPa regime. A prominent structural signature representing the underlying evolution in atomic arrangements and bonding in this pressure regime, as revealed by the ab initio molecular dynamics simulations, is the reduction of low-electron-density regions, which contributes to the narrowing of band gap and delocalization of trapped electrons. At P > 8 GPa, we have observed major changes of the average local structures (bond angle and coordination numbers), gradually transforming the a-GST into a high-density, metallic-like state. This high-pressure glass is characterized by local motifs that bear similarities to the body-centered-cubic GST (bcc-GST) it eventually crystallizes into at 28 GPa, and hence represents a bcc-type polyamorph of a-GST.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Robust resistive memory devices using solution-processable metal-coordinated azo aromatics

Non-volatile memories will play a decisive role in the next generation of digital technology. Flash memories are currently the key player in the field, yet they fail to meet the commercial demands of scalability and endurance. Resistive memory devices, and in particular memories based on low-cost, solution-processable and chemically tunable organic materials, are promising alternatives explored by the industry. However, to date, they have been lacking the performance and mechanistic understanding required for commercial translation. Here we report a resistive memory device based on a spin-coated active layer of a transition-metal complex, which shows high reproducibility (~350 devices), fast switching (≤30 ns), excellent endurance (~10 12 cycles), stability (>10 6 s) and scalability (down to ~60 nm 2 ). In situ Raman and ultraviolet–visible spectroscopy alongside spectroelectrochemistry and quantum chemical calculations demonstrate that the redox state of the ligands determines the switching states of the device whereas the counterions control the hysteresis. This insight may accelerate the technological deployment of organic resistive memories.

Goswami, Sreetosh↗