Characterization of Memory Devices for Energy Efficient Analog In-Memory Neural Computing at the Edge.
Abstract not provided.
Marinella, Matthew↗
SEARCH · Engineering Papers
Results for “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.
Influence of uniaxial compression on the shape memory behavior of vitrimer composite embedded with tension‐programmed unidirectional shape memory polymer fibers
Explore the source record for details and available documents.
Henry Quansah Afful↗
On Nibbles and Bytes: The Conundrum of Memory for Space Systems - NASA Electronic Parts and Packaging (NEPP) and Efforts in Memories
Radiation requirements and trends. TID: 1) >90% of NASA applications are < 100 krads-Si in piecepart requirements. a) Many commercial devices (NVM and SDRAMs) meet or come close to this. b) Charge pump TID tolerance has improved an order magnitude over the last 10 years. 2) There are always a few programs with higher level needs and, of course, defense needs SEL: 1) Prefer none or rates that are considered low risk. a) Latent damage is a bear to deal with. 2) As we re packing cells tighter and even with lower Vdd, we re seeing SEL on commercial devices regularly (<90nm). a) Often in power conversion, I/O, or control areas. SEU: 1) It s not the bit errors, it s the SEFIs errors that are the biggest issues. a) Scrubbing concerns for risk, power, speed.
LaBel, Kenneth A.↗
Development of Stable Two-Way Shape Memory Behavior in a Polycrystalline NiTi Shape Memory Alloy
No abstract available
Benafan, O.↗
Memory technologies in space: radiation hardened magnetic memories
UNKNOWN
Yang-Scharlotta, Jean Y.M.↗
Method and apparatus of integrating memory stacks
A method and apparatus of integrating memory stacks includes providing a first memory die of a first memory technology and a second memory die of a second memory technology. A first logic die is in communication with the first memory die of the first memory technology, and includes a first memory controller including a first memory control function for interpreting requests in accordance with a first protocol for the first memory technology. A second logic die is in communication with the second memory die of the second memory technology and includes a second memory controller including a second memory control function for interpreting requests in accordance with a second protocol for the second memory technology. A memory operation request is received at the first or second memory controller, and the memory operation request is performed in accordance with the associated first memory protocol or the second memory protocol.
Yudanov, Dmitri↗
Blackcomb: Hardware-Software Co-design for Non-Volatile Memory in Exascale Systems
Summary of technical results of Blackcomb Memory Devices We explored various different memory technologies (STTRAM, PCRAM, FeRAM, and ReRAM). The progress can be classified into three categories, below. Modeling and Tool Releases Various modeling tools have been developed over the last decade to help in the design of SRAM or DRAM-based memory hierarchies. To explore new design opportunities that NVM technologies can bring to the designers, we have developed similar high-level models for NVM, including PCRAMsim [Dong 2009], NVSim [Dong 2012], and NVMain [Poremba 2012]. NVSim is a circuit-level model for NVM performance, energy, and area estimation, which supports various NVM technologies, including STT-RAM, PCRAM, ReRAM, and legacy NAND Flash. NVSim is successfully validated against industrial NVM prototypes, and it is expected to help boost architecture-level NVM-related studies. On the other side, NVMain is a cycle accurate main memory simulator designed to simulate emerging nonvolatile memories at the architectural level. We have released these models as open source tools and provided contiguous support to them. We also proposed PS3-RAM, which is a fast, portable and scalable statistical STT-RAM reliability analysis model [Wen 2012]. Design Space Exploration and Optimization With the support of these models, we explore different device/circuit optimization techniques. For example, in [Niu 2012a] we studied the power reduction technique for the application of ECC scheme in ReRAM designs and proposed to use ECC code to relax the BER (Bit Error Rate) requirement of a single memory to improve the write energy consumption and latency for both 1T1R and cross-point ReRAM designs. In [Xu 2011], we proposed a methodology to design STT-RAM for different optimization goals such as read performance, write performance and write energy by leveraging the trade-off between write current and write time of MTJ. We also studied the tradeoffs in building a reliable crosspoint ReRAM array [Niu 2012b]. We have conducted an in depth analysis of the circuit and system level design implications of multi-layer cross-point Resistive RAM (MLCReRAM) from performance, power and reliability perspectives [Xu 2013]. The objective of this study is to understand the design trade-offs of this technology with respect to the MLC Phase Change Memory (MLCPCM).Our MLC ReRAM design at the circuit and system levels indicates that different resistance allocation schemes, programming strategies, peripheral designs, and material selections profoundly affect the area, latency, power, and reliability of MLC ReRAM. Based on this analysis, we conduct two case studies: first we compare MLC ReRAM design against MLC phase-change memory (PCM) and multi-layer cross-point ReRAM design, and point out why multi-level ReRAM is appealing; second we further explore the design space for MLC ReRAM. Architecture and Application We explored hybrid checkpointing using phase-change memory for future exascale systems [Dong 2011] and showed that the use of nonvolatile memory for local checkpointing significantly increases the number of faults covered by local checkpoints and reduces the probability of a global failure in the middle of a global checkpoint to less than 1%. We also proposed a technique called i2WAP to mitigate the write variations in NVM-based last-level cache for the improvement of the NVM lifetime [Wang 2013]. Our wear leveling technique attempts to work around the limitations of write endurance by arranging data access so that write operations can be distributed evenly across all the storage cells. During our intensive research on fault-tolerant NVM design, we found that ECC cannot effectively tolerate hard errors from limited write endurance and process imperfection. Therefore, we devised a novel Point and Discard (PAD) architecture in in [ 2012] as a hard-error-tolerant architecture for ReRAM-based Last Level Caches. PAD improves the lifetime of ReRAM caches by 1.6X-440X under different process variations without performance overhead in the system's early life. We have investigated the applicability of NVM for persistent memory design [Zhao 2013]. New byte addressable NVM enables fast persistent memory that allows in-memory persistent data objects to be updated with much higher throughput. Despite the significant improvement, the performance of these designs is only 50% of the native system with no persistence support, due to the logging or copy-on-write mechanisms used to update the persistent memory. A challenge in this approach is therefore how to efficiently enable atomic, consistent, and durable updates to ensure data persistence that survives application and/or system failures. We have designed a persistent memory system, called Klin, that can provide performance as close as that of the native system. The Klin design adopts a non-volatile cache and a non-volatile main memory for constructing a multi-versioned durable memory system, enabling atomic updates without logging or copy-on-write. Our evaluation shows that the proposed Kiln mechanism can achieve up to 2X of performance improvement to NVRAM-based persistent memory employing write-ahead logging. In addition, our design has numerous practical advantages: a simple and intuitive abstract interface, microarchitecture-level optimizations, fast recovery from failures, and no redundant writes to slow non-volatile storage media. The work was published in MICRO 2013 and received Best Paper Honorable Mentioned Award.
97 MATHEMATICS AND COMPUTING↗
Evaluating the potential of disaggregated memory systems for HPC applications
Summary Disaggregated memory is a promising approach that addresses the limitations of traditional memory architectures by enabling memory to be decoupled from compute nodes and shared across a data center. Cloud platforms have deployed such systems to improve overall system memory utilization, but performance can vary across workloads. High‐performance computing (HPC) is crucial in scientific and engineering applications, where HPC machines also face the issue of underutilized memory. As a result, improving system memory utilization while understanding workload performance is essential for HPC operators. Therefore, learning the potential of a disaggregated memory system before deployment is a critical step. This paper proposes a methodology for exploring the design space of a disaggregated memory system. It incorporates key metrics that affect performance on disaggregated memory systems: memory capacity, local and remote memory access ratio, injection bandwidth, and bisection bandwidth, providing an intuitive approach to guide machine configurations based on technology trends and workload characteristics. We apply our methodology to analyze thirteen diverse workloads, including AI training, data analysis, genomics, protein, fusion, atomic nuclei, and traditional HPC bookends. Our methodology demonstrates the ability to comprehend the potential and pitfalls of a disaggregated memory system and provides motivation for machine configurations. Our results show that eleven of our thirteen applications can leverage injection bandwidth disaggregated memory without affecting performance, while one pays a rack bisection bandwidth penalty and two pay the system‐wide bisection bandwidth penalty. In addition, we also show that intra‐rack memory disaggregation would meet the application's memory requirement and provide enough remote memory bandwidth.
Ding, Nan↗
Latency-agnostic memory controller
A computer-implemented method includes receiving, at a memory controller, a new transaction request referencing a new transaction to be executed on a memory. The memory includes two or more memory groups embodying two or more memory technologies, and the memory controller includes two or more group request queues with a respective group request queue corresponding to each memory group of the two or more memory groups. A memory group is selected, by the memory controller, from among the two or more memory groups. The transaction request is placed, by the memory controller, on the respective group request queue corresponding to the selected memory group. The new transaction is executed on the selected memory group. A new response to the new transaction is received, by the memory controller, from the selected memory group. The new response is returned.
97 MATHEMATICS AND COMPUTING↗
Design Space Exploration of Emerging Memory Technologies for Machine Learning Applications
Memory design space exploration methods study memory systems’ performances and limitations before implementation. The computer memory design space has grown exponentially because of the enormous growth of memory types, memory controllers, and application software. Computer simulators are commonly used for memory design space exploration. However, complex memory simulations take an enormous amount of time. Hence, in this paper, we proposed a machine learning-based design space exploration method for dynamic random-access memory and non-volatile memory systems. We applied our method to the CosmoGAN and LeNet applications to predict the following six memory response parameters: (i) bandwidth, (ii) power, (iii) average latency, (iv) average total latency, (v) memory reads, and (vi) memory writes. Our experimental results show that machine learning models can predict memory response parameter values faster than simulations. We used support vector machine, random forest, and gradient boosting machine learning models. We observed that the support vector machine provides better performance for bandwidth, average latency, and average total latency. The random forest model works better for memory reads and writes. The gradient boosting model provides superior prediction performance for power. We provide a detailed discussion on learning curve characteristics, error analysis, and memory type recommendation.
Hasan, S M Shamimul↗
MOSIQS: Persistent Memory Object Storage With Metadata Indexing and Querying for Scientific Computing
Scientific applications often require high-bandwidth shared storage to perform joint simulations and collaborative data analytics. Shared memory pools provide a chance to satisfy such needs. Recently, a high-speed network such as Gen-Z utilizing persistent memory (PM) offers an opportunity to create a shared memory pool connected to compute nodes. However, there are several challenges to use scientific applications on the shared memory pool directly such as scalability, failure-atomicity, and lack of scientific metadata-based search and query. In this paper, we propose MOSIQS, a persistent memory object storage framework with metadata indexing and querying for scientific computing. We design MOSIQS based on the key idea that memory objects on PM pool can live beyond the application lifetime and can become the sharing currency for applications and scientists. MOSIQS provides an aggregate memory pool atop an array of persistent memory devices to store and access memory objects to accelerate scientific computing. MOSIQS uses a lightweight persistent memory key-value store to manage the metadata of memory objects, which enables memory object sharing. To facilitate metadata search and query over millions of memory objects resident on memory pool, we introduce Group Split and Merge (GSM), a novel persistent index data structure designed primarily for scientific datasets. GSM splits and merges dynamically to minimize the query search space and maintains low query processing time while overcoming the index storage overhead. MOSIQS is implemented on top of PMDK. We evaluate the proposed approach on many-core server with an array of real PM devices. Experimental results show that MOSIQS gains a 100% write performance improvement and executes multi-attribute queries efficiently with 2.7× less index storage overhead offering significant potential to speed up scientific computing applications.
97 MATHEMATICS AND COMPUTING↗
Logic in Memory Emulator
Logic in Memory Emulator (LiME) is a hardware/software tool specially designed for memory system evaluation and experiment. Emerging memories display a wide range of bandwidths, latencies, and capacities, making it challenging for the computer architect to navigate the design space of potential memory configurations, and for the application developer to assess performance implications of using such memories. With the LiME framework, architectural ideas can be prototyped in great detail yet with sufficient performance to support realistic evaluation on long running applications. LiME consists of two fundamental components: 1) the hardware and OS infrastructure for the emulator, and 2) a suite of benchmark applications to assist in characterizing the performance of current and future computer architectures. Some of the applications have been collected from other open source projects. Uses: Logging, replay and analysis of an application's memory behavior Evaluate impact of emerging memory technology on application performance. Emulate complex memory interactions in whole applications orders of magnitude faster than software simulation. Emulate acceleration hardware co-located with the memory subsystem. Features: Capture and log external memory accesses to a separate off-chip memory device without affecting application execution. Memory traces include the address, length, timestamp, and optionally the data for each transaction. Captured trace data can be saved to an SD card for off-line analysis. Configure a wide range of memory latencies in sub-nanosecond increments that encompass highbandwidth and storage class memories. Specify regions of interest (ROI) in applications to reduce the amount of trace data captured for analysis. Currently supports execution on Xilinx Zynq SoC which integrates an ARM processor with FPGA logic on a single device. Applications can be run under Linux or in bare metal mode on the ARM cores.
Jain, AbhishekK↗
Analog Nonvolatile Computer Memory Circuits
In nonvolatile random-access memory (RAM) circuits of a proposed type, digital data would be stored in analog form in ferroelectric field-effect transistors (FFETs). This type of memory circuit would offer advantages over prior volatile and nonvolatile types: In a conventional complementary metal oxide/semiconductor static RAM, six transistors must be used to store one bit, and storage is volatile in that data are lost when power is turned off. In a conventional dynamic RAM, three transistors must be used to store one bit, and the stored bit must be refreshed every few milliseconds. In contrast, in a RAM according to the proposal, data would be retained when power was turned off, each memory cell would contain only two FFETs, and the cell could store multiple bits (the exact number of bits depending on the specific design). Conventional flash memory circuits afford nonvolatile storage, but they operate at reading and writing times of the order of thousands of conventional computer memory reading and writing times and, hence, are suitable for use only as off-line storage devices. In addition, flash memories cease to function after limited numbers of writing cycles. The proposed memory circuits would not be subject to either of these limitations. Prior developmental nonvolatile ferroelectric memories are limited to one bit per cell, whereas, as stated above, the proposed memories would not be so limited. The design of a memory circuit according to the proposal must reflect the fact that FFET storage is only partly nonvolatile, in that the signal stored in an FFET decays gradually over time. (Retention times of some advanced FFETs exceed ten years.) Instead of storing a single bit of data as either a positively or negatively saturated state in a ferroelectric device, each memory cell according to the proposal would store two values. The two FFETs in each cell would be denoted the storage FFET and the control FFET. The storage FFET would store an analog signal value, between the positive and negative FFET saturation values. This signal value would represent a numerical value of interest corresponding to multiple bits: for example, if the memory circuit were designed to distinguish among 16 different analog values, then each cell could store 4 bits. Simultaneously with writing the signal value in the storage FFET, a negative saturation signal value would be stored in the control FFET. The decay of this control-FFET signal from the saturation value would serve as a model of the decay, for use in regenerating the numerical value of interest from its decaying analog signal value. The memory circuit would include addressing, reading, and writing circuitry that would have features in common with the corresponding parts of other memory circuits, but would also have several distinctive features. The writing circuitry would include a digital-to-analog converter (DAC); the reading circuitry would include an analog-to-digital converter (ADC). For writing a numerical value of interest in a given cell, that cell would be addressed, the saturation value would be written in the control FFET in that cell, and the non-saturation analog value representing the numerical value of interest would be generated by use of the DAC and stored in the storage FFET in that cell. For reading the numerical value of interest stored in a given cell, the cell would be addressed, the ADC would convert the decaying control and storage analog signal values to digital values, and an associated fast digital processing circuit would regenerate the numerical value from digital values.
MacLeod, Todd↗
Method and apparatus for faulty memory utilization
A method for faulty memory utilization in a memory system includes: obtaining information regarding memory health status of at least one memory page in the memory system; determining an error tolerance of the memory page when the information regarding memory health status indicates that a failure is predicted to occur in an area of the memory system affecting the memory page; initiating a migration of data stored in the memory page when it is determined that the data stored in the memory page is non-error-tolerant; notifying at least one application regarding a predicted operating system failure and/or a predicted application failure when it is determined that data stored in the memory page is non-error-tolerant and cannot be migrated; and notifying at least one application regarding the memory failure predicted to occur when it is determined that data stored in the memory page is error-tolerant.
97 MATHEMATICS AND COMPUTING↗
UltraLiM: In-Memory Boolean Logic Architecture Using UltraRAM
Conventional computing architectures encounter ‘von Neumann’ and ‘memory wall’ bottlenecks which arise due to the back-and-forth data movement between the physically separate memory and processing units and the speed mismatch between them, respectively. These bottlenecks hurt both energy efficiency and the throughput of computing systems. To address these challenges, in-memory computing architectures have emerged as a promising alternative. They reduce the need for frequent data movement by executing different computing tasks inside the memory system. Here, we present UltraLiM, a logic-in-memory architecture using the UltraRAM-based memory system. UltraRAM holds the promise of developing a ‘universal memory’, overcoming the limitations of charge-based memories thanks to their non-volatile behavior with lower operating voltage. This work presents an in-memory computing architecture that integrates an UltraRAM-based memory array with a custom-designed peripheral circuitry. With this architecture, we can perform various in-memory Boolean logic operations (such as NOT, NAND, NOR, and XOR) in a single cycle. Leveraging the separate read-write paths in the UltraRAM-based memory array, we optimize read operations without encountering design conflicts. This optimization enhances the sense margin, enabling the use of simpler peripheral circuitry for in-memory logic operations.
Alam, Shamiul [University of Tennessee, Knoxville ↗
HAM: Hotspot-Aware Manager for Improving Communications with 3D-Stacked Memory
merging High-Performance Computing (HPC) workloads, such as graph analytics, machine learning, and big data science, are data-intensive. Data-intensive workloads usually present fine-grained memory accesses with limited or no data locality, and thus incur frequent cache misses and low utilization of memory bandwidth. 3D-stacked memory devices such as Hybrid Memory Cube (HMC) and High Bandwidth Memory (HBM) can provide significantly higher bandwidth than conventional memory modules. However, the traditional interfaces and optimization methods for JEDEC DDR devices do not allow to fully exploit the potential performance of 3D-stacked memory with the massive amount of irregular memory accesses of data-intensive applications. In this paper, we propose a novel Hotspot-Aware Manager (HAM) infrastructure for 3D-stacked memory devices capable of optimizing memory access streams via request aggregation, hotspot detection, and in-memory prefetching. %and an associated hotspot-aware page policy. We present the HAM design and implementation, and simulate it on a system using RISC-V embedded cores with attached HMC devices. We extensively evaluate HAM with over 12 benchmarks and applications representing diverse irregular memory access patterns. The results show that, on average, HAM reduces redundant requests by 37.51\% and increases the prefetch buffer hit rate by 4.2 times, compared to a baseline streaming prefetcher. On the selected benchmark set, HAM provides performance gains of 21.81\% in average (up to 34.28\%) and power savings of 35.07\% over a standard 3D-stacked memory.
Wang, Xi↗
Effects of Transcranial Direct Current Stimulation (tDCS) on Human Memory.
Training a person in a new knowledge base or skill set is extremely time consuming and costly, particularly in highly specialized domains such as the military and the intelligence community. Recent research in cognitive neuroscience has suggested that a technique called transcranial direct current stimulation (tDCS) has the potential to revolutionize training by enabling learners to acquire new skills faster, more efficiently, and more robustly (Bullard et al., 2011). In this project, we tested the effects of tDCS on two types of memory performance that are critical for learning new skills: associative memory and working memory. Associative memory is memory for the relationship between two items or events. It forms the foundation of all episodic memories, so enhancing associative memory could provide substantial benefits to the speed and robustness of learning new information. We tested the effects of tDCS on associative memory, using a real-world associative memory task: remembering the links between faces and names. Working memory refers to the amount of information that can be held in mind and processed at one time, and it forms the basis for all higher-level cognitive processing. We investigated the degree of transfer between various working memory tasks (the N-back task as a measure of verbal working memory, the rotation-span task as a measure of visuospatial working memory, and Raven's progressive matrices as a measure of fluid intelligence) in order to determine if tDCS-induced facilitation of performance is task-specific or general.
Matzen, Laura E.↗