Kernel Modification Research Articles

Tuning high-level synthesis SpMV kernels in Alveo FPGAs

Sparse Matrix-Vector Multiplication (SpMV) is an essential operation in scientific and engineering fields, with applications in areas like finite element analysis, image processing, and machine learning. To address the need for faster and more energy-efficient computing, this paper investigates the acceleration of SpMV through Field-Programmable Gate Arrays (FPGAs), leveraging High-Level Synthesis (HLS) for design simplicity. Our study focuses on the AMD-Xilinx Alveo U280 FPGA, assessing the performance of the SpMV kernel from Vitis Libraries, which is the state of the art on SpMV acceleration on FPGAs. We explore kernel modifications, transition to single precision, and varying partition sizes, demonstrating the impact of these changes on execution time. Furthermore, we investigate matrix preprocessing techniques, including Reverse Cuthill-McKee (RCM) reordering and a hybrid sparse storage format, to enhance efficiency. Our findings reveal that the performance of FPGA-accelerated SpMV is influenced by matrix characteristics, by smaller partition sizes, and by specific preprocessing techniques delivering notable performance improvements. By selecting the best results from these experiments, we achieved execution time enhancements of up to 3.2×. This study advances the understanding of FPGA-accelerated SpMV, providing insights into key factors that impact performance and potential avenues for further improvement.

Solutions to the Balitsky-Kovchegov equation including the dipole orientation

Solutions of the target-rapidity Balitsky-Kovchegov (BK) equation are studied considering, for the first time, the complete impact-parameter dependence, including the orientation of the dipole with respect to the impact-parameter vector. In our previous work [1] it has been demonstrated that the spurious Coulomb tails could be tamed using the collinearly-improved kernel and an appropriate initial condition in the projectile-rapidity BK equation. Introducing a different interpretation of the evolution variable, the target-rapidity formulation of the BK equation brings non-locality in rapidity and a kernel modification, removing the term that previously helped to suppress the Coulomb tails. To address this newly emerged non-locality, three different prescriptions are explored here to take into account the rapidities preceding the initial condition. Two of these approaches induce mild Coulomb tails, while the other is free from this effect within the studied rapidity range. The range is chosen to correspond to that of interest for existing and future experiments. To demonstrate that this set up can be used for phenomenological studies, the obtained solutions are used to compute the F2 structure function of the proton and the diffractive photo- and electro-production of J/ψ off protons. The predictions agree well with HERA data, confirming that the target-rapidity Balitsky-Kovchegov equation with the full impact-parameter dependence is a viable tool to study the small Bjorken-x limit of perturbative QCD at current facilities like RHIC and LHC as well as in future colliders like the EIC.

Controlled Shell and Kernel Modifications of Atomically Precise Pd/Ag Superatomic Nanoclusters.

The first 8-electron Pd/Ag superatomic alloys with an interstitial hydride [PdHAg19(dtp)12] (dtp = S2P(O i Pr)2-) 1 and [PdHAg20(dtp)12]+2 are reported. The targeted addition of a single Ag atom to 1 is achieved by the reaction of one equivalent of trifluoroacetic acid, resulting in the formation of 2 in 55% yield. Further modification of the shell results in the formation of [PdAg21(dtp)12]+3 via an internal redox reaction, with the system retaining an 8-electron superatomic configuration. The interstitial hydride in 1 and 2, contributes its 1s1 electron to the superatomic electron count, and occupies a PdAg3 tetrahedron. The distributions of isomers corresponding to different dispositions of the outer capping Ag atoms are investigated by multinuclear VT-NMR spectroscopy. The emissive state of 3 has a lifetime of 200 μs (λex =448; λem =842), while 1 and 2 are non-emissive. The catalytic reduction of 4-nitrophenol is demonstrated with 1-3 at room temperature.

Using rootkits hiding techniques to conceal honeypot functionality

Honeypot is one of the existing technologies in the area of computer network security. The goal of Honeypot is to create a tempting target for the attacker. The system that is considered as a Honeypot in the network includes the services and functions of a real system that the attacker sees as a normal system and enters for exploitation. In this way, Honeypot can monitor the behaviors, patterns, and tools used in various attacks. Certainly, if the intelligent attacker realizes the existence of this trap in the target network, he can design ways to bypass it. In this case, Honeypot practically loses its effectiveness. Therefore, the issue of hiding Honeypot is one of the primary challenges in this field. In this paper, a solution to this problem is presented. In the present paper, Honeypot is concealed using the concealment techniques used in Rootkits. We use application examples and theoretical analysis results to show that the proposed Honeypots concealment approach is strong against existing kernel-based Honeypots detection methods. Sebek, VMScope, and Qebek are three Honeypots that we choose for comparison purposes. The proposed Honeypot has been compared with them in virtualization, memory usage, and kernel modification. The experimental results show that the proposed hidden Honeypot in addition to low kernel modification has no track in the memory. Also, the proposed Honeypot does not use virtualization. It can successfully be concealed in the kernel of the target system without any effect on the target system. We also implemented the proposed algorithm on the example network and test all Sebek’s detection methods on it. The experimental results show that the proposed kernel-based approach can bypass these detection methods.

Performance analysis of real-time and general-purpose operating systems for path planning of the multi-robot systems

In general, modern operating systems can be divided into two essential parts, real-time operating systems (RTOS) and general-purpose operating systems (GPOS). The main difference between GPOS and RTOS is the system istime-critical or not. It means that; in GPOS, a high-priority thread cannot preempt a kernel call. But, in RTOS, a low-priority task is preempted by a high-priority task if necessary, even if it’s executing a kernel call. Most Linux distributions can be used as both GPOS and RTOS with kernel modifications. In this study, two Linux distributions, Ubuntu and Pardus, were analyzed and their performances were compared both as GPOS and RTOS for path planning of the multi-robot systems. Robot groups with different numbers of members were used to perform the path tracking tasks using both Ubuntu and Pardus as GPOS and RTOS. In this way, both the performance of two different Linux distributions in robotic applications were observed and compared in two forms, GPOS, and RTOS.

Rcryptect: Real-time detection of cryptographic function in the user-space filesystem

The existing methods of ransomware detection have limitations. To be specific, static analysis is not effective to obfuscated binaries, while dynamic analysis is usually restricted to a certain platform and often takes tens of minutes. In this paper, we propose a block-level monitoring system to detect potentially malicious cryptographic operations. We carry out statistical analysis to find heuristic rules to distinguish between normal and encrypted blocks. In order to apply the heuristic rule to the filesystem without kernel modification, we adopt Filesystem in Userspace (FUSE) and define our filesystem Rcryptect for real-time detection of cryptographic function. We demonstrate the protection of well-known ransomware and show that various cryptographic functions can be detected with about 13% overhead.

Random radial basis function kernel-based support vector machine

The main computational cost of building a support vector machine (SVM) training model lies in tuning the hyperparameters, including the kernel parameters and penalty constant C. This paper introduces a new kernel, the random radial basis function (RRBF) kernel, which all kernel parameters can be assigned to randomly. The key idea of the RRBF is to extend a one-dimensional parameter to d-dimensional parameters by simple modification of the traditional RBF kernel. We prove the universal approximation capability of a SVM with the RRBF kernel, proposing a simple model selection algorithm. The experiments on benchmark datasets show that SVM with the RRBF kernel clearly outperforms the traditional RBF kernel and other popular kernels, and the results are quite insensitive to C. Compared with other kernels, a SVM with the RRBF kernel is not sensitive to the penalty constant C, and it can be tuned in a wide range. Additionally, this property also enables a SVM to find the hyperparameters quickly. A simulation of parameter sensitivity study also show that a SVM is more stable w.r.t. to the different choices of kernel parameters.

Complex Kernel-based spectrum reconstruction algorithm for cascaded Fabry-Perot interferometric spectrometer.

A method to calculate the spectrum of the light incident on a cascaded Fabry-Perot interferometric spectrometer from the detector signal versus the scanning mirror position is presented. The method is based on modifying the Fabry-Perot integral equation to reduce possible spectrum reconstruction errors that arise due to inaccurate determination of the optical path difference reference position and the dependence on the dispersion of the cavity material. A transformation algorithm that employs the suggested kernel modification is derived and tested. The presented algorithm is then compared to the conventional kernel, showing spectral error reductions by larger than 20 dB.

Non-Volatile Main Memory Emulator for Embedded Systems Employing Three NVMM Behaviour Models

Emerging byte-addressable non-volatile memory devices attract much attention. A non-volatile main memory (NVMM) built on them enables larger memory size and lower power consumption than a traditional DRAM main memory. To fully utilize an NVMM, both software and hardware must be cooperatively optimized. Simultaneously, even focusing on a memory module, its micro architecture is still being developed though real non-volatile memory modules, such as Intel Optane DC persistent memory (DCPMM), have been on the market. Looking at existing NVMM evaluation environments, software simulators can evaluate various micro architectures with their long simulation time. Emulators can evaluate the whole system fast with less flexibility in their configuration than simulators. Thus, an NVMM emulator that can realize flexible and fast system evaluation still has an important role to explore the optimal system. In this paper, we introduce an NVMM emulator for embedded systems and explore a direction of optimization techniques for NVMMs by using it. It is implemented on an SoC-FPGA board employing three NVMM behaviour models: coarse-grain, fine-grain and DCPMM-based. The coarse and fine models enable NVMM performance evaluations based on extensions of traditional DRAM behaviour. The DCPMM-based model emulates the behaviour of a real DCPMM. Whole evaluation environment is also provided including Linux kernel modifications and several runtime functions. We first validate the developed emulator with an existing NVMM emulator, a cycle-accurate NVMM simulator and a real DCPMM. Then, the program behavior differences among three models are evaluated with SPEC CPU programs. As a result, the fine-grain model reveals the program execution time is affected by the frequency of NVMM memory requests rather than the cache hit ratio. Comparing with the fine-grain model and the coarse-grain model under the condition of the former's longer total write latency than the latter's, the former shows lower execution time for four of fourteen programs than the latter because of the bank-level parallelism and the row-buffer access locality exploited by the former model.

Sintering conditions recognition of rotary kiln based on kernel modification considering class imbalance

Accurate sintering condition recognition (SCR) is an important precondition for optimal control of rotary kilns. However, the occurrence probability of abnormal conditions in the industrial field is much lower than normal, resulting in imbalanced class sintering samples in general. This significantly deteriorates the effectiveness of existing recognition models in abnormal condition detection. In this paper, an integrated framework considering class imbalance is proposed for sintering condition recognition. In the proposed framework, after analysing the characteristics of thermal signals by the Lipschitz method, four discriminant features are extracted to comprehensively describe different sintering conditions. In addition, focusing on the class imbalance of sintering samples, the kernel modification method is introduced to enhance the optimal marginal distribution machine (ODM), and a novel recognition model kernel modified the ODM (KMODM) is proposed for SCR. By constructing a new conformal transformation function to modify the ODM kernel function, KMODM optimizes the spatial distribution of training samples in the kernel space, thereby alleviating the detection accuracy deterioration of the minority class. The experimental results on real thermal signals and standard datasets show that the KMODM model can effectively handle imbalanced data. Based on this, the proposed SCR framework can reduce the misjudgement of abnormal conditions and balance the recognition accuracy of each condition.

A Sintering State Recognition Framework to Integrate Prior Knowledge and Hidden Information Considering Class Imbalance

Estimation of the sintering state has importance for clinker quality improvements and the safe operation of the rotary kiln. Class imbalanced thermal signals usually pose challenges in feature extraction and abnormal state recognition. In this article, a novel framework that integrates prior knowledge and hidden information is developed for sintering state recognition in the class imbalance condition. For discriminative feature extraction of imbalanced data, a cascaded stack autoencoder (SAE) model is proposed to fuse our prior knowledge and hidden information. The model includes a feature extraction SAE and a deep fusion SAE: the former extracts hidden information from thermal signals, and the latter deeply fuses and compresses our prior knowledge and hidden information. For the class imbalance of sintering samples, we propose a data-dependent kernel modification optimal margin distribution machine (ddKMODM) as a sintering state recognition model. Modifying the original kernel function by a conformal function depending on the data distribution in kernel space, ddKMODM can change the local volume expansion coefficient of the feature space to eliminate the negative effects caused by imbalanced samples. Experiments on real data show that the proposed framework can balance the detection rate of each state in the class imbalanced condition, and its overall sintering state recognition accuracy exceeds 92%.

Screening length for finite-size ions in concentrated electrolytes.

The classical Debye-Hückel (DH) theory clearly accounts for the origin of screening in electrolyte solutions and works rather well for dilute electrolyte solutions. While the Debye screening length decreases with the ion concentration and is independent of ion size, recent surface-force measurements imply that for concentrated solutions, the screening length exhibits an opposite trend; it increases with ion concentration and depends on the ionic size. The screening length is usually defined by the response of the electrolyte solution to a test charge but can equivalently be derived from the charge-charge correlation function. By going beyond DH theory, we predict the effects of ion size on the charge-charge correlation function. A simple modification of the Coulomb interaction kernel to account for the excluded volume of neighboring ions yields a nonmonotonic dependence of the screening length (correlation length) on the ionic concentration, as well as damped charge oscillations for high concentrations.

Modification of the agglomeration kernel and simulation of the flow pattern in acoustic field with fine particles

This paper studies the acoustic agglomeration of fine particles of the coal fly ash in the acoustic field agglomeration device, based on the coupling population balance model (PBM) for acoustic agglomeration of fine particles and the modified kinetic theory of the fine particle flow. The new root mean square of the agglomeration kernel function operates according to the acoustic agglomeration mechanism. We present the theoretical investigation of the problem and then compare it with experimental results. The results show that the agglomeration process of particles in different size ranges is dominated by different agglomeration mechanisms. Moreover, we show the numerical simulation of the acoustic agglomeration process of the particles using the new agglomeration kernel function, which is consistent with the experimental results. In addition, the particle flow in the acoustic field agglomeration device presents a typical turbulent structure, and a large particle agglomeration occurs near the wall and outlet.

A high-resolution gravimetric quasigeoid model for Vietnam

A high-resolution gravimetric quasigeoid model for Vietnam and its surrounding areas was determined based on new gravity data. A set of 29,121 land gravity measurements was used in combination with fill-in data where no gravity data existed. Global Gravity field Models plus Residual Terrain Model effects and gravity field derived from altimetry satellites were used to provide the fill-in information over land and marine areas. A mixed model up to degree/order 719 was used for the removal of the long and medium wavelengths and the calculation of the quasigeoid restore effects. The residual height anomalies have been determined employing the Stokes integral using the Fast Fourier Transform approach and deterministic kernel modification proposed by Wong–Gore, as well as by means of Least-Squares Collocation. The accuracy of the resulting quasigeoid models was evaluated by comparing with height anomalies derived from 812 co-located GNSS/levelling points. Results are very similar; both local quasigeoid models have a standard deviation of 9.7 cm and 50 cm in mean bias when compared to the GNSS/levelling points. This new local quasigeoid model for Vietnam represents a significant improvement over the global models EIGEN-6C4 and EGM2008, which have standard deviations of 19.2 and 29.1 cm, respectively, for this region.

Aurophilic Interactions in the Self-Assembly of Gold Nanoclusters into Nanoribbons with Enhanced Luminescence.

Aurophilic interactions (AuI ⋅⋅⋅AuI ) are crucial in directing the supramolecular self-assembly of many gold(I) compounds; however, this intriguing chemistry has been rarely explored for the self-assembly of nanoscale building blocks. Herein, we report on studies on aurophilic interactions in the structure-directed self-assembly of ultrasmall gold nanoparticles or nanoclusters (NCs, <2 nm) using [Au25 (SR)18 ]- (SR=thiolate ligand) as a model cluster. The self-assembly of NCs is initiated by surface-motif reconstruction of [Au25 (SR)18 ]- from short SR-[AuI -SR]2 units to long SR-[AuI -SR]x (x>2) staples accompanied by structure modification of the intrinsic Au13 kernel. Such motif reconstruction increases the content of AuI species in the protecting shell of Au NCs, providing the structural basis for directed aurophilic interactions, which promote the self-assembly of Au NCs into well-defined nanoribbons in solution. More interestingly, the compact structure and effective aurophilic interactions in the nanoribbons significantly enhance the luminescence intensity of Au NCs with an absolute quantum yield of 6.2 % at room temperature.

Aurophilic Interactions in the Self‐Assembly of Gold Nanoclusters into Nanoribbons with Enhanced Luminescence

AbstractAurophilic interactions (AuI⋅⋅⋅AuI) are crucial in directing the supramolecular self‐assembly of many gold(I) compounds; however, this intriguing chemistry has been rarely explored for the self‐assembly of nanoscale building blocks. Herein, we report on studies on aurophilic interactions in the structure‐directed self‐assembly of ultrasmall gold nanoparticles or nanoclusters (NCs, &lt;2 nm) using [Au25(SR)18]− (SR=thiolate ligand) as a model cluster. The self‐assembly of NCs is initiated by surface‐motif reconstruction of [Au25(SR)18]− from short SR‐[AuI‐SR]2 units to long SR‐[AuI‐SR]x (x&gt;2) staples accompanied by structure modification of the intrinsic Au13 kernel. Such motif reconstruction increases the content of AuI species in the protecting shell of Au NCs, providing the structural basis for directed aurophilic interactions, which promote the self‐assembly of Au NCs into well‐defined nanoribbons in solution. More interestingly, the compact structure and effective aurophilic interactions in the nanoribbons significantly enhance the luminescence intensity of Au NCs with an absolute quantum yield of 6.2 % at room temperature.

A Nonlocal Isoperimetric Problem with Dipolar Repulsion

We study a geometric variational problem for sets in the plane in which the perimeter and a regularized dipolar interaction compete under a mass constraint. In contrast to previously studied nonlocal isoperimetric problems, here the nonlocal term asymptotically localizes and contributes to the perimeter term to leading order. We establish existence of generalized minimizers for all values of the dipolar strength, mass and regularization cutoff and give conditions for existence of classical minimizers. For subcritical dipolar strengths we prove that the limiting functional is a renormalized perimeter and that for small cutoff lengths all mass-constrained minimizers are disks. For critical dipolar strength, we identify the next-order $\Gamma$-limit when sending the cutoff length to zero and prove that with a slight modification of the dipolar kernel there exist masses for which classical minimizers are not disks.

Accurate Ranging on Acoustic-Enabled IoT Devices

The enabling Internet-of-Things technology has inspired many innovative sensing mechanisms by repurposing the onboard sensors. Leveraging the built-in acoustic sensors for ranging is among one of the interesting applications. However, among the few studies on acoustic ranging, the one-way sensing method suffers from synchronization errors and requires cumbersome kernel modifications; the other two-way approaches overcome these shortcomings, but they are sensitive to system delays. In this case, this paper proposes a novel lightweight one-way sensing paradigm without the above drawbacks. The key insight of this paper is to perform ranging by estimating the propagation time of acoustic signals via linear frequency modulation signal mixing. Such a signal mix operation can translate range estimation into fine-grain frequency estimation, thereby enhancing ranging accuracy. In addition, our system can have multiple receivers co-exist and thus the measurement dimensions are boosted. We have implemented and evaluated our system prototype in real-world settings. The prototype demonstrated centimeter-level ranging performance.

Generalised Born-Infeld models, Lax operators and the mathrm{T}overline{mathrm{T}} perturbation

Surprising links between the deformation of 2D quantum field theories induced by the composite mathrm{T}overline{mathrm{T}} operator, effective string models and the AdS/CFT correspondence, have recently emerged. The purpose of this article is to discuss various classical aspects related to the deformation of 2D interacting field theories. Special attention is given to the sin(h)-Gordon model, for which we were able to construct the mathrm{T}overline{mathrm{T}} -deformed Lax pair. We consider the Lax pair formulation to be the first essential step toward a more satisfactory geometrical interpretation of this deformation within the integrable model framework.Furthermore, it is shown that the 4D Maxwell-Born-Infeld theory, possibly with the addition of a mass term or a derivative-independent potential, corresponds to a natural extension of the 2D examples. Finally, we briefly comment on 2D Yang-Mills theory and propose a modification of the heat kernel, for a generic surface with genus p and n boundaries, which fully accounts for the mathrm{T}overline{mathrm{T}} contribution.

Unified estimation of densities on bounded and unbounded domains

Kernel density estimation in domains with boundaries is known to suffer from undesirable boundary effects. We show that in the case of smooth densities, a general and elegant approach is to estimate an extension of the density. The resulting estimators in domains with boundaries have biases and variances expressed in terms of density extensions and extension parameters. The result is that they have the same rates at boundary and interior points of the domain. Contrary to the extant literature, our estimators require no kernel modification near the boundary and kernels commonly used for estimation on the real line can be applied. Densities defined on the half-axis and in a unit interval are considered. The results are applied to estimation of densities that are discontinuous or have discontinuous derivatives, where they yield the same rates of convergence as for smooth densities on $${\mathbb {R}}$$ .