Full Memory Research Articles

Full Spatial and Temporal Memory Safety for C

Full Spatial and Temporal Memory Safety for C

An Efficient Limited Memory Multi-Step Quasi-Newton Method

This paper is dedicated to the development of a novel class of quasi-Newton techniques tailored to address computational challenges posed by memory constraints. Such methodologies are commonly referred to as “limited” memory methods. The method proposed herein showcases adaptability by introducing a customizable memory parameter governing the retention of historical data in constructing the Hessian estimate matrix at each iterative stage. The search directions generated through this novel approach are derived from a modified version closely resembling the full memory multi-step BFGS update, incorporating limited memory computation for a singular term to approximate matrix–vector multiplication. Results from numerical experiments, exploring various parameter configurations, substantiate the enhanced efficiency of the proposed algorithm within the realm of limited memory quasi-Newton methodologies category.

Effectiveness of Online Learning in English Subject at Pangkalpinang state Junior High School during Covid-19 Pandemic

Effective online learning is virtual face-to-face learning carried out through communication and information technology by teachers and students since the emergence of the Covid-19 pandemic by fulfilling 5 indicators (management of learning implementation, communicative processes, student responses, learning activities and learning outcomes) which are supported by adequate infrastructure (Internet network, smartphone memory, internet quota) especially English language learning. Is the implementation of online English learning effective at Pangkalpinang State Middle School? remembering that there were several obstacles found in the field, namely difficult networks, full smartphone memory, insufficient internet quota, bad weather such as thunderstorms, and the absence of students' parents to accompany them during the learning process. This research is a mix-method research that uses a questionnaire distributed via WhatsApp as a data collection method. The questionnaire contains 8 closed questions and 3 open questions. The population contains all state junior high school students in Pangkalpinang. Samples were taken using cluster random sampling. The data analysis technique used is descriptive statistical analysis, the results of which are developed using qualitative analysis.

Soft error mitigation and recovery of SRAM-based FPGAs using brain-inspired hybrid-grained scrubbing mechanism.

Soft error has increasingly become a critical concern for SRAM-based field programmable gate arrays (FPGAs), which could corrupt the configuration memory that stores configuration data describing the custom-designed circuit architecture. To mitigate this kind of error, this study proposes a brain-inspired hybrid-grained scrubbing mechanism consisting of fine-grained and coarse-grained scrubbing to mitigate and repair the errors as quickly as possible after an SEU occurrence. Inspired by the human brain's ability to filter out redundant and irrelevant information, we propose a mechanism that can mask invalid position information when errors occur. Compared with the scrubbing of full configuration memory, this mechanism can achieve precise error location and recovery utilizing targeted scrubbing of specific frames or modules. The effectiveness is evaluated by executing fault injection campaigns on the International Symposium on Circuits and Systems 1989 (ISCAS89) benchmark circuits and fault tolerant fast Fourier transform (FT-FFT) circuit. If upsets are detected, they will be repaired with fine-grained or coarse-grained scrubbing depending on their location. The experiment results show that this mechanism can effectively mitigate and repair single-bit upsets (SBUs) and double-bit upsets (DBUs). In addition, the mechanism is proven to be superior in error recovery time and hardware overhead compared to counterpart approaches.

Full communication memory networks for team-level cooperation learning

Full communication memory networks for team-level cooperation learning

ADRA: Extending Digital Computing-In-Memory With Asymmetric Dual-Row-Activation

Computing in-memory (CiM) has emerged as an attractive technique to mitigate the von-Neumann bottleneck. Current digital CiM approaches for in-memory operands are based on multi-wordline assertion for computing bit-wise Boolean functions and arithmetic functions such as addition. However, most of these techniques, due to the many-to-one mapping of input vectors to bitline voltages, are limited to CiM of commutative functions, leaving out an important class of computations such as subtraction. In this paper, we propose a CiM approach, which solves the mapping problem through an asymmetric wordline biasing scheme, enabling (a) simultaneous single-cycle memory read and CiM of primitive Boolean functions (b) computation of any Boolean function and (c) CiM of non-commutative functions such as subtraction and comparison. While the proposed technique is technology-agnostic, we show its utility for ferroelectric transistor (FeFET)-based non-volatile memory. Compared to the standard near-memory methods (which require two full memory accesses per operation), we show that our method can achieve a full scale two-operand digital CiM using just one memory access, leading to a 23.2% -72.6% decrease in energy-delay product (EDP).

Comparative analysis of three memory selection methods for time integration of Fractional Reaction-Diffusion Equations

By discretising in space, a non-linear time fractional reaction-diffusion equations (TFRDEs) can be converted into a system of time-fractional differential equations (TFDEs). The full memory method (FMM) and short memory method (SMM) are well-established memory selection methods used in the time integration of TFDEs. The main drawbacks of FMM and SMM are higher computational cost and uncontrollable error respectively. The only way to increase the accuracy of SMM is by increasing short memory length which causes an increase in computational cost. Especially when we apply these two methods to integrate TFRDEs, we have to solve a large system of TFDEs. Therefore, the drawbacks of these two methods affect seriously, when these are applied to solve TFRDEs. This paper aims to investigate the accuracy and efficiency of the memory selection method, Exponentially Decreasing Random Memory Method (EDRMM), and compare it with FMM and SMM when these methods apply to integrate TFRDEs. Based on these three memory selection methods, three semi-implicit numerical schemes namely semiimplicit scheme with full memory method (SI-FMM), semi-implicit scheme with short memory method (SI-SMM), and semi-implicit scheme with exponentially decreasing random memory method (SI-EDRMM)) are proposed and the accuracy and CPU time (computational time (CT)) of these three numerical schemes are compared. To do this comparison, these three numerical schemes are applied to four TFRDEs whose exact solutions are known. Numerical experiments confirm that the accuracy and efficiency of the SI-EDRMM are better than that of SI-SMM and the efficiency of SI-EDRMM is higher than that of SI-FMM. Therefore, EDRMM is better than SMM and FMM for the integration of TFRDEs.

Full shape memory effect of Cu-13.5Al-4Ni-6Fe shape memory martensite single crystal

The Cu−13.5Al−4Ni−6Fe centimeter-leveled martensite single crystals with full shape memory effect were prepared through simply annealing the cast alloy. The microstructure, reversible martensitic transformation, shape memory effect of this single crystal were investigated using optical microscopy, transmission electron microscope, differential scanning calorimeter, thermal mechanical analysis and compression test. The results show that Cu−13.5Al−4Ni−6Fe single crystal consists of 18R and 2H martensite, as well as rich-Fe bcc nanoprecipitate. The starting temperatures and the finishing temperatures of the forward and reverse martensitic transformations are 387, 397, 365 and 432 K, respectively. When the pre-strain is 6%, 7%, and 8%, this single crystal exhibits full shape memory effect of 4.4%, 4.7% and 5.2%, respectively. The obtained Cu−13.5Al−4Ni−6Fe martensite single crystal possesses shape memory properties that can be comparable to those of traditional shape memory alloys.

Cerebras Architecture Deep Dive: First Look Inside the Hardware/Software Co-Design for Deep Learning

The compute and memory demands for deep learning and machine learning (ML) have increased by several orders of magnitude in just the last couple of years, and there is no end in sight. Traditional improvements in processor performance alone struggle to keep up with the exponential demand. A new chip architecture co-designed with the ML algorithms can be better equipped to satisfy this unprecedented demand and enable the ML workloads of the future. This article describes the Cerebras architecture and how it is designed specifically with this purpose, from the ground up, as a wafer-sized chip to enable emerging extreme-scale ML models. It uses fine-grained data flow compute cores to accelerate unstructured sparsity, distributed static random-access memory for full memory bandwidth to the data paths, and a specially designed on-chip and off-chip interconnect for ML training. With these techniques, the Cerebras architecture provides unique capabilities beyond traditional designs.

Improving the Capacity of Quantum Dense Coding and the Fidelity of Quantum Teleportation by Weak Measurement and Measurement Reversal.

A protective scheme of quantum dense coding and quantum teleportation of the X-type initial state is proposed in amplitude damping noisy channel with memory using weak measurement and measurement reversal. Compared with the noisy channel without memory, the memory factor improves both the capacity of quantum dense coding and the fidelity of the quantum teleportation to a certain extent for the given damping coefficient. Although the memory factor can inhibit decoherence in some degree, it cannot eliminate it completely. In order to further overcome the influence of the damping coefficient, the weak measurement protective scheme is proposed, which found that the capacity and the fidelity can be efficiently improved by adjusting weak measurement parameter. Another practical conclusion is that, among the three initial states, the weak measurement protective scheme has the best protective effect on the Bell-state in terms of the capacity and the fidelity. For the channel with no memory and full memory, the channel capacity of quantum dense coding reaches two and the fidelity of quantum teleportation reaches one for the bit system; the Bell system can recover the initial state completely with a certain probability. It can be seen that the entanglement of the system can be well protected by the weak measurement scheme, which provides a good support for the realization of quantum communication.

Pemanfaatan Learning Management System (LMS) Berbasis Moodle dalam Pembelajaran Sejarah Pada Kelas X SMK Negeri 3 Magelang

The purpose of this study is to know the use of LMS Moodle in history study and to know of the obstacles encountered by teachers and trainees in making use of the LMS Moodle. The study uses a qualitative approach with case study strategies. Data collection techniques of observation, interview, and document review. The data collected will then be processed to inference. Research shows that during learning done from home 1) 1) history study with an LMS Moodle is done well. The planning, implementation and evaluations of the history teacher have been systematic. The casting of learning with the LMS Moodle was presensi, matter waiver, and evaluation by taking an formative question. During the online study the history teacher has made good use of, but in doing so has not been fully effective. 2) the challenges teachers experience are the making of attractive materials, the media selection to explain the material, the teachers cannot control the learning activities, and the activation of the learners. The performance of learning is inadequate because of age factors, and learners' accessibility is still lacking, and the class's organization is irregular. The challenge for a learner is the often defective LMS Moodle, an unstable Internet network, a lack of understanding of matter, and a usually full cell phone memory.

Intermediary roles of prospective memory and retrospective memory in the comorbidity of depression and pain

BackgroundPatients who suffer comorbidity of major depressive disorder (MDD) and chronic pain (CP) maintain a complex interplay between maladaptive prospective memory (PM) and retrospective memory (RM) with physical pain, and...

MSWasm: Soundly Enforcing Memory-Safe Execution of Unsafe Code

Most programs compiled to WebAssembly (Wasm) today are written in unsafe languages like C and C++. Unfortunately, memory-unsafe C code remains unsafe when compiled to Wasm—and attackers can exploit buffer overflows and use-after-frees in Wasm almost as easily as they can on native platforms. Memory- Safe WebAssembly (MSWasm) proposes to extend Wasm with language-level memory-safety abstractions to precisely address this problem. In this paper, we build on the original MSWasm position paper to realize this vision. We give a precise and formal semantics of MSWasm, and prove that well-typed MSWasm programs are, by construction, robustly memory safe. To this end, we develop a novel, language-independent memory-safety property based on colored memory locations and pointers. This property also lets us reason about the security guarantees of a formal C-to-MSWasm compiler—and prove that it always produces memory-safe programs (and preserves the semantics of safe programs). We use these formal results to then guide several implementations: Two compilers of MSWasm to native code, and a C-to-MSWasm compiler (that extends Clang). Our MSWasm compilers support different enforcement mechanisms, allowing developers to make security-performance trade-offs according to their needs. Our evaluation shows that on the PolyBenchC suite, the overhead of enforcing memory safety in software ranges from 22% (enforcing spatial safety alone) to 198% (enforcing full memory safety), and 51.7% when using hardware memory capabilities for spatial safety and pointer integrity. More importantly, MSWasm’s design makes it easy to swap between enforcement mechanisms; as fast (especially hardware-based) enforcement techniques become available, MSWasm will be able to take advantage of these advances almost for free.

Fractional-order integral generalization of the Rapoport–Leas equation

A one-dimensional time-fractional integral model for a two-phase flow through a porous medium with power-law memory in the presence of capillary forces is considered. Based on the time-fractional integral generalization of Darcy’s law and the classical mass balance equations for phases, neglecting gravity forces, a time-fractional integral generalization of the Rapoport–Leas equation is derived. It is shown that in the limiting case of constant capillary pressure, the obtained equation coincides with the classical Buckley–Leverett equation. An example of initial boundary value problem for the derived fractional integral equation is given for the case of a finite reservoir. Unlike the classical Rapoport–Leas equation, its time-fractional integral analogue does not have a traveling wave solution in general case. It is due to the fact that the fractional integral with a finite lower limit is not invariant with respect to translations in time. However, for the model problem of two-phase displacement in an unbounded reservoir, considered as a discontinuity breakup problem, the full memory approximation is valid. In this case the obtained equation has a traveling wave solution. In this approximation, relations for the main characteristics of the saturation jump on the shock are obtained. In particular, it is shown that the speed of the shock can be found from an equation that includes the time-fractional integral characteristic of the Leverett function. The behavior of the shock near its left and right boundaries is also asymptotically studied for different saturation levels at the boundaries. It has been proven that if the initial saturation exceeds the residual saturation of the displacing phase, then on the right boundary the saturation at the jump exponentially tends to the initial one. On the left boundary, the asymptotic behavior of the jump is also exponential, with the exception of one special case in which it becomes a fractional power law. For the equation in question, an approximate solution has been analytically constructed for the case of «weak» full memory, when the order of fractional integration is close to zero and can be considered as a small parameter.

Russia-Ukraine war: independence, identity, and security

The aim of this short paper is to show that the relations between Ukraine and Russia are extremely complex as they are built on a shared history, religion, language, and culture and they should not, therefore, be gauged by Western standards. Evaluating the current situation from a broader perspective, the fact that Ukraine plays a significant role in Russia’s foreign policy needs to be emphasized. Ukraine is considered the key post-Soviet state, a significant “near abroad” country, whose position, potential, and geopolitical location are vital for the balance of power in both Eastern Europe and Europe in general. The research project aims at a synthetic examination of the current Russia-Ukraine war from the point of view of Ukraine striving for full independence, memory, and identity. Will the war of the brotherly nations wreck Russia’s long-term policy towards Ukraine? Will the conflict ruin Russia’s plans of post-Soviet space reintegration, whose pivot is set on Ukraine? Will the war stimulate Ukraine to deal with the issue of its identity, post-Soviet legacy, and collective memory?

A novel generalized fractional‐order memristor model with fully explicit memory description

AbstractIn this work, a novel generalized mathematical model of fractional‐order memristor with fully explicit memory description has been proposed. For obtaining such full explicit memory description, the Atangana‐Baleanu fractional derivative in Liouville‐Caputo sense, which employs a nonsingular kernel, has been adopted as the mathematical basis. The proposed model has been derived without regarding to any specific conventional memristor. A comparison with the singular kernel fractional derivative‐based model has been made. The behavioral analysis of the fractional‐order memristor based on the proposed model has been performed, where both DC and AC stimuli have been considered. In addition, its application to the practical fractional‐order memristor‐based circuit and its extension to the fractional‐order memreactance have also been shown. Unlike the singular kernel fractional derivative‐based model, a fully explicit memory description can be obtained by ours. Many other interesting results that are contradict to the previous singular kernel fractional derivative‐based ones, e.g., the fractional‐order memristor that can be locally active, have been demonstrated. The abovementioned extension can be conveniently performed. In summary, this is the first time that a nonsingular kernel fractional derivative has been applied to the fractional‐order memristor modeling and the resulting model with a fully explicit memory description has been proposed. The proposed model is also highly generic, applicable to the practical circuit, and extendable to the fractional‐order memreactance.

PEMANFAATAN APLIKASI WHATSAPP GROUP (WAG) SEBAGAI SARANA PEMBELAJARAN JARAK JAUH (PJJ) PADA SISWA KELAS IV (Studi Kasus Pembelajaran di Era Pandemi di SDN Kencana Indah II)

This study aims to describe the use of the WhatsApp Group Application as a Distance Learning Media for Class IV Students at SDN Kencana Indah II, the obstacles that occur in the use of the WhatsApp Group Application as a Distance Learning Media for Class IV Students at SDN Kencana Indah II and the solutions implemented to overcome these obstacles. This study uses descriptive research with a qualitative approach. In this study, the subjects were Teachers, Students, and Parents of Class IV Students. The research data was collected through observation, interviews, questionnaires, and documentation studies. The results of the study indicate that teachers have used the WhatsApp Group application as a learning medium in supporting distance learning activities by utilizing various available features such as photo, video, document, telephone, and video call features. The ease is felt by the presence of the WhatsApp Group feature and also its easy use and can be used by various groups in teaching and learning activities, this is supported by the results of a questionnaire which states that 55% of respondents strongly agree with the ease of use of the WhatsApp Group application as a learning medium and 39% of respondents thers agree with the statement. Then in the implementation of the use of WhatsApp Group as a distance learning medium, there are several obstacles, such as signal interference, full cellphone memory, lack of interaction, difficulty knowing the seriousness of students' learning, lack of student motivation to learn, online learning support facilities, and difficulty understanding the material provided, as well as in this study describe the solutions carried out to overcome the obstacles that occur from the use of the WhatsApp Group application as a distance learning medium.

Breaking Cuckoo Hash: Black Box Attacks

Introduced less than twenty years ago, cuckoo hashing has a number of attractive features like a constant worst case number of memory accesses for queries and close to full memory utilization. Cuckoo hashing has been widely adopted to perform exact matching of an incoming key with a set of stored (key, value) pairs in both software and hardware implementations. This widespread adoption makes it important to consider the security of cuckoo hashing. Most hash based data structures can be attacked by generating collisions that reduce their performance. In fact, for cuckoo hashing collisions could lead to insertion failures which in some systems would lead to a system failure. For example, if cuckoo hashing is used to perform Ethernet lookup and a given MAC address cannot be added to the cuckoo hash, the switch would not be able to correctly forward frames to that address. Previous works have shown that this can be done when the attacker knows the hash functions used in the implementation. However, in many cases the attacker would not have that information and would only have access to the cuckoo hash operations to perform insertions, removals or queries. This article considers the security of a cuckoo hash to an attacker that has only a black box access to it. The analysis shows that by carefully performing user operations on the cuckoo hash, the attacker can force insertion failures with a small set of elements. The proposed attack has been implemented and tested for different configurations to demonstrate its feasibility. The fact that cuckoo hash can be broken with only access to its user functions should be taken into account when implementing it in critical systems. The article also discusses potential approaches to mitigate this vulnerability.

A Stochastic Momentum Accelerated Quasi-Newton Method for Neural Networks (Student Abstract)

Incorporating curvature information in stochastic methods has been a challenging task. This paper proposes a momentum accelerated BFGS quasi-Newton method in both its full and limited memory forms, for solving stochastic large scale non-convex optimization problems in neural networks (NN).

Analisis Hasil Belajar Matermatika Dalam Pembelajaran Daring Menggunakan Aplikasi Whatsapp Group

This study aims to describe the results of learning mathematics during the online learning process and the obstacles in using the whatsapp group application in fourth grade elementary school students.This research is a qualitative research with the type of narrative research which was carried out in SD 2 Mlati Lor, Kota District, Kudus Regency. The subjects in this study were 6 students, 3 boys and 3 girls. Data collection techniques used observation, interviews and documentation. Data analysis used is data reduction, data presentation, and drawing conclusions. The results of the research obtained are that online learning using the Whatsapp Group application at SD 2 Mlati Lor runs well according to the applicable educational standards.The results showed that all educational units were on average in online learning using the Whatsapp Group Application. The results of learning mathematics at SD 2 Mlati Lor during online learning are good, as indicated by the complete score of all students. There are obstacles in the use of Whatsapp Group as a learning medium at SD 2 Mlati Lor, namely signal interference, lack of motivation, inadequate facilities, difficulty in understanding the material, full mobile memory and too many tasks.