Random Access Memory Module Research Articles

ML-driven risk estimation for memory failure in a data center environment with convolutional neural networks, self-supervised data labeling and distribution-based model drift determination

With the trend towards multi-socket server systems, the demand for random access memory (RAM) per server increased. The consequence are more DIMM sockets per server. Since every dual in-line memory module (DIMM), which comprises a series of dynamic random-access memory integrated circuits, has a probability of failure, RAM issues became a dominant failure pattern for servers. The concept introduced in this work contributes to improving the reliability of data centers by avoiding RAM failures and mitigating their impact. For this purpose, an ML-driven framework is provided to estimate the probability of memory failure for each RAM module. The ML framework is based on structural information between correctable (CE) and uncorrectable errors (UE). In a common memory scenario, a corrupted bit within a module can be restored by redundancy using an error correction code (ECC), resulting in a CE. However, if there is more than one corrupted bit within a group of bits covered by the ECC, the information cannot be restored, resulting in a UE.Consequently, the related task requesting the memory content, and the corresponding service may crash. There is evidence that UEs have a CE history and structural relation between the CEs. However, for the case of UEs without a CE history or of a false decision of the ML framework, we extend the total framework by engineering measures to mitigate the impact of a UE by avoiding kernel panic and using backups. The engineering measures use a mapping between physical and logical memory addresses.

Risk assessment for sustainability in e-waste recycling in circular economy

E-waste is a hazardous concept for human health, environment and businesses, and therefore, risk assessment is essential to eliminate or reduce the negative effects of e-waste in a circular economy. Decision-making plays an important role during the risk assessment in e-waste recycling. This study aims to develop a decision tool under uncertain and risky conditions for achieving sustainability in electronic waste (e-waste) recycling in circular economy. Prospect theory and its aspects are integrated with the sustainability risk assessment in e-waste recycling. In order to evaluate the irrationality and risk attitudes of decision makers in a risky and uncertain environment, the prospect theory-based TODIM (acronym in Portuguese for interactive and multicriteria decision-making, i.e. Tomada de Decisao Interativa Multicriterio) method for a previous computer disassembly problem is used. Using prospect theory, the relationship between loss aversion, reference point, framing and two selves effect, and sustainability dimensions is presented with a sustainability risk assessment point of view. Further, tasks of computer disassembly processes are prioritized by considering risk factors. The results showed that random access memory modules, motherboard and power supply are the highest priority risks based on the evaluation of listed twelve sustainability risks. In this study, it is suggested that prospect theory is suitable for decisions under risk and uncertainty, and it is applicable to integrate sustainability risk assessment and prospect theory in order to fulfil the gap of knowledge. Relationship between prospect theory and sustainability risk assessment.

Hybrid OCS/OBS interconnect in intra-data-center network

We report on a data center network (DCN) architecture based on hybrid optical circuit switching (OCS) and optical burst switching (OBS) interconnect for dynamic DCN connectivity provisioning. With the combination of the centralized and distributed control of the software-defined optical networks, the proposed interconnect can achieve unprecedented flexibility in dealing with both mice and elephant flow in the DCN. Numerical simulation is employed to investigate the performance of the proposed architecture. The results show that the OBS module has preferable performance in dealing with a larger burst packet, and the throughput is constrained by the capacity of the server random access memory module.

The methodology for active testing of electronic devices under the radiations

The methodology, developed for active testing of electronic devices under the radiations, is presented. The test set-up includes a gamma-ray facility, the hardware board/fixtures and the software tools purposely designed and realized. The methodology is so wide-ranging to allow us the verification of different classes of electronic devices, even if only application examples for static random access memory modules are reported.

Spincaloric properties of epitaxialCo2MnSi/MgO/Co2MnSimagnetic tunnel junctions

The electronic transport and spincaloric properties of epitaxial magnetic tunnel junctions with half-metallic ${\mathrm{Co}}_{2}\mathrm{MnSi}$ Heusler electrodes, MgO tunneling barriers, and different interface terminations are investigated by using first-principles calculations. An approach to spincaloric properties is presented that circumvents the linear response approximation inherent in the Seebeck coefficient and compared to the method of Sivan and Imry. This approach supports two different temperatures in the two electrodes and provides the exact current and/or voltage response of the system. Moreover, it accounts for temperature-dependent chemical potentials in the electrodes and finite-bias effects. We find that especially the former are important for obtaining qualitatively correct results, even if the variations of the chemical potentials are small. It is shown how the spincaloric properties can be tailored by the choice of the growth conditions. We find a large effective and spin-dependent Seebeck coefficient of $\ensuremath{-}65\phantom{\rule{4pt}{0ex}}\ensuremath{\mu}\mathrm{V}/\mathrm{K}$ at room temperature for the purely Co-terminated interface. We suggest to use such interfaces in thermally operated magnetoresistive random access memory modules, which exploit the magneto-Seebeck effect, to maximize the thermally induced readout voltage.

Specific energy consumption of dynamic random access memory module supply chain in Taiwan

This study focuses on specific energy consumption (SEC) of all major energy consuming segments of the Dynamic Random Access Memory (DRAM) module supply chain, including 12” Si-wafer (ingot), wafer fabrication, integrated circuit (IC) assembly, IC testing, module assembly, and printed circuit board (PCB). Results show that energy consumption of a single unit of the end product from the DRAM supply chain is 4.67 kWh/PIC (PIC = a piece of DRAM module). For its annual shipments of 85,136,366 PIC/yr, the whole DRAM supply chain consumes a total of 397,331,726 kWh of energy, emitting 252,703 tons of CO2. The outcomes of this study provide baseline SEC of the DRAM supply chain that can be used while designing fabrication plants (hereafter referred to as “fabs”) in the future. We estimate that if all fabs in Taiwan achieve the same level of SEC as the best performer (i.e. lowest SEC level), annual energy savings and emissions reduction in Taiwan could be expected to reach 228,561,645 kWh of electricity and 149,426 tons of CO2 emissions, respectively.

3D laser trimming technology for regulating embedded thick-film carbon resistors on a random access memory module

This study presents a novel solution for regulating embedded thick-film carbon resistors on random access memory (RAM) modules using 3D laser trimming technology. The embedded passive components, which are printed on a substrate using liquid-polymer ink, are employed due to their rapid response time. The original resistance of such resistors, however, is less than the target value. Hence, a 3D laser trimming technology is proposed for etching thick-film resistors and reaching a target resistance value with online assistance of a four-wire resistance measurement device. A laser beam deflection path with a spiral cycle is manufactured on the printed thick-film to superficially thin the resistor using a wobble-grinding function in a layer-by-layer manner. Resistance material was removed by each laser pulse until the target resistance was obtained. Twelve RAM modules were tested. Experimental results demonstrate that the resistance value of the trimmed resistors was an averaged accredited percentage of 88.54% with a tolerance of 5%. All trimmed resistors with a tolerance of 10% were obtained. Furthermore, 144,000 embedded thick-film carbon resistors on 1800 RAM modules were trimmed in a factory using the proposed scheme. Measurement results indicate that an average accredited rate of 95.67% with tolerances of 5–20% was obtained for industrial applications. Therefore, the proposed technology is a practical and efficient solution for regulating thick-film carbon resistors on RAM modules and other related products.

A parallel pipelined relational query processor

This paper presents the design of a relational query processor. The query processor consists of only four processing PIPEs and a number of random-access memory modules. Each PIPE processes tuples of relations in a bit-serial, tuple-parallel manner for each of the primitive database operations which comprise a complex relational query. The design of the query processor meets three major objectives: the query processor must be manufacturable using existing and near-term LSI (VLSI) technology; it must support in a uniform manner both the numeric and nonnumeric processing requirements a high-level user interface like SQL presents; and it must support the query-processing strategy derived in the query optimizer to satisfy certain system-wide performance optimality criteria.

A parallel pipelined relational query processor

This paper outlines the overall architecture of a query processor for relational queries and describes the design and control of its major processing modules. The query processor consists of only four processing modules and a number of random-access memory modules. Each processing module processes tuples of relations in a bit-serial, tuple-parallel manner for each of the primitive database operations which comprise a complex relational query. The query processor is designed to be manufacturable using existing VLSI technology, and to support in a uniform manner both the numeric and nonnumeric processing requirements a high-level query language like SQL presents.

Bandwidth of fast memory in multiprocessing

A Markov chain analysis is presented for evaluating the average memory bandwidth of a multiple microprocessor system where the random access memory modules have a cycle time shorter than the processor cycle time. The results are useful in the design of such systems, as well as in the understanding of the effects of such processor-memory speed mismatches.

Potential of superconductive josephson tunneling technology for ultrahigh performance memories and processors

The potential of a new technology, based on the superconducting Josephson tunneling effect, has been assessed for use in high-speed computer systems. Josephson tunneling circuits for memory and logic functions can be switched at sub-nanosecond delay, do not require standby power, and dissipate extremely low energy, typically less than 10-13joule during fast switching operation. High-speed circuits with high packing densities, not limited by power dissipation problems, can therefore be expected. Low thermal noise at cryogenic temperatures ensures reliable operation despite the low energy switching signal. Practically lossless signal transmission via superconducting lines can be exploited for high-speed operation even when sizeable memory and logic modules are assembled. Two random access memory modules have been designed on the basis of single device experiments. Operational characteristics leading to the conclusion that 30 Mbit capacity, less than 1 watt refrigeration requirement (at 3.6° K) and cycle times of 40 ns (in one case) and 15 ns (in the second case), can be achieved are presented.