Loop Splitting Research Articles

Competing magnetic states and M–H loop splitting in core–shell NiO nanoparticles

Magnetic relaxation in a nanoparticles system depends on the intra-particle interactions, reversal mechanism, the anisotropy field, easy axis distribution, particle volume, lattice defects, surface defects, materials composite, etc. Here we report the competing magnetic states between superparamagnetic blocking and Néel transition states in 14 nm core–shell NiO nanoparticles. A crossover temperature of 50 K was observed for both these states from the zero field cooled/field cooled magnetization curves taken at different fields. At crossover temperature, an interesting M–H loop splitting is observed which is attributed to the slow spin relaxation. This anomalous M–H loop splitting behaviour was found to be particle size dependent and suppressed for diameters above and below 14 nm which indicates a critical size for these competing magnetic states. Additional neutron diffraction experiments confirmed this observation. This experimental study provides a new insight for the understanding of intra-particle interactions in fine antiferromagnetic nanoparticles and obtained results are an important step towards deeper understanding of the competing/non-competing modes between superparamagnetic blocked and Néel transition states.

Composite topological nodal lines penetrating the Brillouin zone in orthorhombic AgF2

It has recently been found that nonsymmorphic symmetries can bring many exotic band crossings. Here, based on symmetry analysis, we predict that materials with time-reversal symmetry in the space group of Pbca (No. 61) possess rich symmetry-enforced band crossings, including nodal surfaces, fourfold degenerate nodal lines and hourglass Dirac loops, which appear in triplets as ensured by the cyclic permutation symmetry. We take Pbca AgF2 as an example in real systems and studied its band structures with ab initio calculations. Specifically, in the absence of spin-orbit coupling (SOC), besides the above-mentioned band degeneracies, this system features a nodal chain and a nodal armillary sphere penetrating the Brillouin zone (BZ). While with SOC, we find a new configuration of the hourglass Dirac loop/chain with the feature traversing the BZ, which originates from the splitting of a Dirac loop confined in the BZ. Furthermore, guided by the bulk-surface correspondence, we calculated the surface states to explore these bulk nodal phenomena. The evolution of these interesting nodal phenomena traversing the BZ under two specific uniaxial strains is also discussed.

Bridge trisections of knotted surfaces in 4-manifolds

We prove that every smoothly embedded surface in a 4-manifold can be isotoped to be in bridge position with respect to a given trisection of the ambient 4-manifold; that is, after isotopy, the surface meets components of the trisection in trivial disks or arcs. Such a decomposition, which we call a generalized bridge trisection, extends the authors' definition of bridge trisections for surfaces in [Formula: see text] Using this construction, we give diagrammatic representations called shadow diagrams for knotted surfaces in 4-manifolds. We also provide a low-complexity classification for these structures and describe several examples, including the important case of complex curves inside [Formula: see text] Using these examples, we prove that there exist exotic 4-manifolds with [Formula: see text]-trisections for certain values of g. We conclude by sketching a conjectural uniqueness result that would provide a complete diagrammatic calculus for studying knotted surfaces through their shadow diagrams.

MULTI-PARAMETRIC STUDY OF RISING 3D BUOYANT FLUX TUBES IN AN ADIABATIC STRATIFICATION USING AMR

We study the buoyant rise of magnetic flux tubes embedded in an adiabatic stratification using two-and three-dimensional, MHD simulations. We analyze the dependence of the tube evolution on the field line twist and on the curvature of the tube axis in different diffusion regimes. To be able to achieve a comparatively high spatial resolution we use the FLASH code, which has a built-in Adaptive Mesh Refinement (AMR) capability. Our 3D experiments reach Reynolds numbers that permit a reasonable comparison of the results with those of previous 2D simulations. When the experiments are run without AMR, hence with a comparatively large diffusivity, the amount of longitudinal magnetic flux retained inside the tube increases with the curvature of the tube axis. However, when a low-diffusion regime is reached by using the AMR algorithms, the magnetic twist is able to prevent the splitting of the magnetic loop into vortex tubes and the loop curvature does not play any significant role. We detect the generation of vorticity in the main body of the tube of opposite sign on the opposite sides of the apex. This is a consequence of the inhomogeneity of the azimuthal component of the field on the flux surfaces. The lift force associated with this global vorticity makes the flanks of the tube move away from their initial vertical plane in an antisymmetric fashion. The trajectories have an oscillatory motion superimposed, due to the shedding of vortex rolls to the wake, which creates a Von Karman street.

Model of exchange bias in a trilayer FM/AFM/FM structure

A theoretical examination of the properties of the exchange bias phenomenon for sandwich FM/AFM/FM structures, which are actively studied in experiments. Magnetization dependences on the external magnetic field are obtained under the proposed discrete model of ferromagnetic layers of such multilayer structures. It is shown that the magnetization field dependences have a horizontal plateau, a splitting of the hysteresis loop, and asymmetry. We examine the influence of the interface defects on the exchange bias phenomenon and its features. The results are qualitatively consistent with experimental data.

Spin-orbit interaction driven collective electron-hole excitations in a noncentrosymmetric nodal loop Weyl semimetal

NbP is one member of a new class of nodal loop semimetals characterized by the cooperative effects of spin-orbit coupling (SOC) and a lack of inversion center. Here transport and spectroscopic properties of NbP are evaluated using density functional theory methods. SOC together with the lack of inversion symmetry splits degeneracies, giving rise to "Russian doll nested" Fermi surfaces containing 4*10$^{-4}$ electron (hole) carriers/f.u. Due to the modest SOC strength in Nb, the Fermi surfaces map out the Weyl nodal loops. Calculated structure around T$^*$~100 K in transport properties reproduces well the observed transport behavior only when SOC is included, attesting to the precision of the (delicate) calculations and the stoichiometry of the samples. Low energy collective electron-hole excitations (plasmons) in the 20-60 meV range result from the nodal loop splitting.

A loop splitting method for single loops with non-uniform dependences

Computationally expensive programs spend most of their time in the execution of DO-loops [1,2]. Therefore, an efficient approach for exploiting potential parallelism is to concentrate on the parallelism available in loops in ordinary programs and has a considerable effect on the speedup [3,4]. Through some loop transformations using a dependence distance in single loops [6,10], a loop can be splitted into partial loops to be executed in parallel without violating data dependence relations, that is, the size of distance can be used as a reduction factor [6], which is the number of iterations that can be executed in parallel. In the case of non-constant distance such that it varies between different instances of the dependence, it is much more difficult to maximize the degree of parallelism from a loop. Partitioning of loops requires efficient and exact data dependence analysis. A precise dependence analysis helps in identifying dependent/independent iterations of a loop. And it is important that appropriate dependence analysis be applied to exploit maximum parallelism within loops. We can consider some tests that examine the dependence of onedimensional subscripted variables— the separability test, the GCD test and the Banerjee test [8,11]. In general, the GCD test is applied first because of its simplicity, even if it is an approximate test. Next, for the case that the gcd test is true, the separability test is attempted again, and through this exact

Finger Vein Representation by Modified Binary Tree Model

Finger vein recognition has high identification accuracy and strong security performance, which can be used in banks, offices, factories, etc. Although image representation is not a necessary process for finger vein recognition, a proper representation method can help to explore distribution regularities and structure differences of finger veins, and provides instructive information for finger vein recognition. It is very difficult to represent finger veins because of their irregular structure. Therefore, four principles (caliber uniformity, node replication, loop splitting, and virtual connection) are proposed in this paper, first to simplify the finger vein structure as a binary tree structure. Then a modified binary tree model is proposed based on the binary tree structure. The new model uses the binary tree to describe the relationships between different vein branches and uses a B-spline function to describe the spatial structure of vein branches. Experiments show that this model can quantitatively describe the relationships between, and the spatial structure of, vein branches with little representation error and low storage space requirements.

공유 메모리 병렬 프로그램의 수행중 오류 탐지를 위한 루프 분리

병렬 프로그램은 의도되지 않은 비결정적인 수행을 야기하므로 공유 메모리를 사용하는 병렬 프로그램에서는 경합을 탐지하는 것은 매우 중요하다. 수행 중 기법에서 경합을 탐지하기 위해서 요구되는 기억장소의 부담은 매우크다. 특히 동기화가 있는 병렬 프로그램에서 경합 탐지에 필요한 기억 공간의 문제는 더욱 심각하다. 그래서, 본 논문에서는 원시 프로그램의 시멘틱을 유지하면서 동기화를 가지는 공유 메모리 병렬 프로그램의 디버깅을 위한 루프 분리 기법을 제시한다. 이것은 동기화를 가지는 병렬 프로그램의 수행 중 경합 탐지에 필요로 하는 기억공간의 복잡성을 줄일 수 있고, 루프 분리된 프로그램을 수행 중에 감시하여 최초 경합들을 탐지할 수 있다. Detecting races is important for debugging shared-memory parallel programs, because the races result in unintended non-deterministic executions of the programs. Previous on-the-fly techniques to detect races in parallel programs with general inter-thread coordination show serious space overhead which depends on the maximum parallelism of the program. Therefore, this paper presents a loop splitting technique for on-the-fly race detection of parallel programs which is more efficient in space complexity than previous techniques. This loop splitting technique is the debugged program which preserves semantics of the original program. Monitering loop splitting program in on-the-fly can detect first races.

On threshold resummation of singlet structure and fragmentation functions

The large- x behavior of the physical evolution kernels appearing in the second order evolution equations of the singlet F 2 structure function and of the F ϕ structure function in ϕ-exchange DIS is investigated. The validity of a leading logarithmic threshold resummation, analogous to the one prevailing for the non-singlet physical kernels, is established, allowing to recover the predictions of Soar et al. for the double-logarithmic contributions ( ln i ( 1 − x ) , i = 4 , 5 , 6 ) to the four loop splitting function P q g ( 3 ) ( x ) and P g q ( 3 ) ( x ) . Threshold resummation at the next-to-leading logarithmic level is found however to break down in the three loop kernels, except in the “supersymmetric” case C A = C F . Assuming a full threshold resummation does hold in this case also beyond three loop gives some information on the leading and next-to-leading single-logarithmic contributions ( ln i ( 1 − x ) , i = 2 , 3 ) to P q g ( 3 ) ( x ) and P g q ( 3 ) ( x ) . Similar results are obtained for singlet fragmentation functions in e + e − annihilation up to two loop, where a large- x Gribov–Lipatov relation in the physical kernels is pointed out. Assuming this relation also holds at three loop, one gets predictions for all large- x logarithmic contributions to the three loop timelike splitting function P g q ( 2 ) T ( x ) , which are related to similar terms in P q g ( 2 ) ( x ) .

Compiler-directed memory management for heterogeneous MPSoCs

Advances in semiconductor technique enable multiple processor cores to be integrated into a single chip. Heterogeneous multiprocessor system-on-a-chip (MPSoC) becomes important platforms to accelerate applications. However, compilation techniques for memory management on MPSoCs still lag behind. This paper presents an automatic memory management framework to orchestrate the data movement between local memory and off-chip memory. In our framework, data alignment, hierarchically data distribution, communication generation, loop tiling, and loop splitting are employed. Moreover, a communication optimization approach is proposed to improve data reuse. These techniques can reduce off-chip memory access and exploit data locality. Experimental results on Cell BE show that our data management framework can generate efficient code for the program.

Parallelism for Nested Loops with Simple Subscripts

In this paper, we propose improved loop splitting method for maximizing parallelism of single loops with non-constant dependence distances. By using the iteration and distance for the source of the first dependence, and by our defined theorems, we present generalized and optimal algorithms for single loops with non-uniform dependences (MPSL). By the extension of the MPSL method, we also apply to exploit parallelism from nested loops with simple subscripts, based on cycle shrinking and loop interchanging method. The algorithms generalize how to transform general single loops with non-uniform dependences as well as nested loops with simple subscripts into parallel loops.

Partial data reuse for nested loop computations: design space exploration for FPGA implementations

Automated compiler analyses and transformation techniques aim at improving design productivity of the mapping process of applications expressed in high-level programming languages to FPGAs. These transformations allow a compiler tool to reduce the number of design cycles and eliminate tedious and error-prone low-level transformations required in this mapping process, while still leading to good designs. Scalar replacement, also known as register promotion, is a very important data-oriented transformation that leads to designs that reduce the number of external memory accesses, and thus reduce execution time, at the expense of storage resource's. In this article we present a combination of loop transformation techniques, namely loop unrolling, loop splitting, and loop interchange with scalar replacement to enable partial data reuse on computations expressed by tightly nested loops pervasive in image processing algorithms. We describe a performance modelling in the presence of partial data reuse. Our experimental results reveal that our model accurately captures the non-trivial execution effects of pipelined implementations in the presence of partial data reuse due to the need to fill-up data buffers. The model thus allows a compiler to explore a large design space with high accuracy, ultimately allowing quickly it to find better designs than designs with limited manual search or brute-force approaches.

Parallelism for Single Loops with Multiple Dependences

We review some loop partitioning techniques such as loop splitting method by thresholds and Polychronopoulos' loop splitting method for exploiting parallelism from single loop which already developed. We propose improved loop splitting method for maximizing parallelism of single loops with non-constant dependence distances. By using the iteration and distance for the source of the first dependence, and by our defined theorems, we present generalized and optimal algorithms for single loops with non-uniform dependences. The algorithms generalize how to transform general single loops with one dependence as well as with multiple dependences into parallel loops.

Finite element analysis of thixoforming process with arbitrary shape die

Numerical simulations for industrial net shape forming application by the finite element method are based on the integral form of the mechanical equations governing the physical process. However, current commercial packages cannot analyze the die filling and solidification for materials in the semi-solid state because they lack material models for liquid and solid materials and do not address segregation of the liquid and solid phases. Thus, this work presents an overview of the development of a new finite element program entitled SEMI-FORM S/W to predict the solid-phase deformation of semi-solid materials (SSM), filling defects and liquid segregation. The automatic mesh generation scheme by the modified looping algorithm using quadrilateral elements is proposed in this present study by a new loop splitting method and a local finite element control. An automatic quadrilateral mesh generator with automatic mesh refinement is developed in order to meet special demands found in the thixoforming process analysis with arbitrary shaped dies by the node renumbering and modifications of the nodal position. The remeshing due to B-spline form is carried out by applying the remeshing criterion. The SEMI-FORM S/W developed in this present study would be recognized as efficient and economical tool for designing and optimizing thixoformed products with arbitrary shaped dies.

Improving workload balance and code optimization on processor-in-memory systems

Processor-in-memory (PIM) architectures have recently been proposed, with the objective of reducing the performance gap between processor and memory. An earlier study of Huang and Chu [Proceedings of 2nd Workshop on Intelligent Memory Systems, Cambridge, MA, 2000] designed a statement-based parallelizing system, SAGE, to exploit the potential benefits of PIM. This study extends this system to achieve better performance. Several comprehensive optimization approaches, including self-patch weight evaluation, loop splitting for PIM, intelligent memory operation (IMOP) recognition, and tiling for PIM, are devised to produce execution schedules with improved load balance. Experimental results confirm the effectiveness of the proposed method.

Forging process analysis with arbitrary shape die in semi-solid material with high solid fraction

Numerical simulations for industrial net shape forming application by the finite element method (FEM) are based on the integral form of the mechanical equations governing the physical process. However, current commercial packages cannot analyze the die filling and solidification for materials in the semi-solid state because they lack material models for liquid and solid materials and do not address segregation of the liquid and solid phases. Thus, this work presents an overview of the development of a new finite element program entitled SEMI-FORM S/W to predict the solid-phase deformation of semi-solid materials, filling defects and liquid segregation. The automatic mesh generation scheme by the modified looping algorithm using quadrilateral elements is proposed in this present study by a new loop splitting method and a local finite element control. An automatic quadrilateral mesh generator with automatic mesh refinement is developed in order to meet special demands found in the thixoforming process analysis with arbitrarily shaped dies by the node renumbering and modifications of the nodal position. Applying the remeshing criterion carries out the remeshing due to B-spline form. The SEMI-FORM S/W developed in this present study would be recognized as efficient and economical tool for designing and optimizing thixoformed products with arbitrarily shaped dies.

Generation of Efficient Nested Loops from Polyhedra

Automatic parallelization in the polyhedral model is based on affine transformations from an original computation domain (iteration space) to a target space-time domain, often with a different transformation for each variable. Code generation is an often ignored step in this process that has a significant impact on the quality of the final code. It involves making a trade-off between code size and control code simplification/optimization. Previous methods of doing code generation are based on loop splitting, however they have nonoptimal behavior when working on parameterized programs. We present a general parameterized method for code generation based on dual representation of polyhedra. Our algorithm uses a simple recursion on the dimensions of the domains, and enables fine control over the tradeoff between code size and control overhead.

Simplifying control flow in compiler-generated parallel code

Optimizing compilers for data-parallel languages such as High Performance Fortran perform a complex sequence of transformations. However, the effects of many transformations are not independent, which makes it challenging to generate high quality code. In particular, some transformations introduce conditional control flow, while others make some conditionals unnecessary by refining program context. Eliminating unnecessary conditional control flow during compilation can reduce code size and remove a source of overhead in the generated code. This paper describes algorithms to compute symbolic constraints on the values of expressions used in control predicates and to use these constraints to identify and remove unnecessary conditional control flow. These algorithms have been implemented in the Rice dHPF compiler and we show that these algorithms are effective in reducing the number of conditionals and the overall size of generated code. Finally, we describe a synergy between control flow simplification and data-parallel code generation based on loop splitting which achieves the effects of more narrow data-parallel compiler optimizations such as vector message pipelining and the use of overlap areas.

Nonlinear finite element analysis of beams and arches using parallel processors

A nonlinear finite element formulation for beams and arches is presented for implementation on a multiple instruction multiple data (MIMD) parallel machine. A twelve degree of freedom arch element is developed using Green strains and 2nd Piola Kirchhoff stresses. Concepts like loop splitting and expression splitting are used to improve the load balancing capability among the processors. Using this developed element, beams and arches that are undergoing large displacements and large rotations are analyzed, and efficiency of the program is compared with a sequential implementation.