Parallelism Constraints Research Articles

Improving Interference Analysis for Real-Time DAG Tasks Under Partitioned Scheduling

Real-time systems with strict timing constraints have been widely applied in many fields. The Directed acyclic graph (DAG) task model has been widely studied and applied to model real-time systems with partial parallelism and precedence constraints in each task. Our paper focuses on the worst-case response time (WCRT) analysis of DAG tasks under partitioned scheduling on multiprocessors. We investigate a parallel structure named <formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex>$Str$</tex> </formula> , which helps obtain more accurate analysis results, and propose a new offline scheduling analysis algorithm named reducing repetitive calculation (RRC). Experiments with synthetic workload are conducted to compare the results calculated by RRC and the state-of-the-art, as well as the observed average response time on a real embedded system. Results show that RRC has better performance in terms of analysis accuracy.

Singularity Analysis for the Perspective-Four and Five-Line Problems

This paper deals with image-based visual servoing and pose estimation by observing four and five lines. Our main interest is to determine the relative configurations of the camera and the observed lines that lead to problems in control and stability. Since it is equivalent to finding the singularities of the corresponding Jacobian matrix, we use tools from computational algebraic geometry to seek configurations such that all of its minors vanish simultaneously. By choosing a suitable basis for this matrix, we revisit the problem in the case of three lines to show that one type of the singularities is when the camera lies on the hyperboloid of one sheet uniquely defined by the lines. This result is further exploited to prove that the one-dimensional singularities, if any, in the case of n lines appear when the camera lies on the transversals to the observed lines. Thus, by forcing the transversals to be complex, we can avoid the aforementioned type of singularities in the case of four lines although the algebra shows that there can always be up to 10 inevitable singular locations of the camera for the other type of singularity. For five lines, we find out that there are no singularities in the generic case. The singularities are also characterized for four and five lines with orthogonality and parallelism constraints. Furthermore, a visual servoing library is used to conduct some simulated experiments to substantiate the theoretical results. As expected, we observe problems in control in the vicinity of a singularity as well as increased errors in pose estimation.

Processing gapping: Parallelism and grammatical constraints.

This study aims to test two hypotheses about the online processing of Gapping: whether the parser inserts an ellipsis site in an incremental fashion in certain coordinated structures (the Incremental Ellipsis Hypothesis), or whether ellipsis is a late and dispreferred option (the Ellipsis as a Last Resort Hypothesis). We employ two offline acceptability rating experiments and a sentence fragment completion experiment to investigate to what extent the distribution of Gapping is controlled by grammatical and extra-grammatical constraints. Furthermore, an eye-tracking while reading experiment demonstrated that the parser inserts an ellipsis site incrementally but only when grammatical and extra-grammatical constraints allow for the insertion of the ellipsis site. This study shows that incremental building of the Gapping structure follows from the parser's general preference to keep the structure of the two conjuncts maximally parallel in a coordination structure as well as from grammatical restrictions on the distribution of Gapping such as the Coordination Constraint.

QMLE

Estimating the demands placed by services on physical resources is an essential step for the definition of performance models. For example, scalability analysis relies on these parameters to predict queueing delays under increasing loads. In this article, we investigate maximum likelihood (ML) estimators for demands at load-independent and load-dependent resources in systems with parallelism constraints. We define a likelihood function based on state measurements and derive necessary conditions for its maximization. We then obtain novel estimators that accurately and inexpensively obtain service demands using only aggregate state data. With our approach, and also thanks to approximation methods for computing marginal and joint distributions for the load-dependent case, confidence intervals can be rigorously derived, explicitly taking into account both topology and concurrency levels of the services. Our estimators and their confidence intervals are validated against simulations and real system measurements for two multi-tier applications, showing high accuracy also in models with load-dependent resources.

A new approach to predict the flexibility and precision of manufacturing systems using geometric constraints and small displacement torsors

Cost and quality of a product are prime drivers in manufacturing. It is important to optimize these variables for maximum benefit. It is to be noted that higher quality means a product which has fewer errors. The quality of product is associated with the precision of the manufacturing system but machines to produce such products are expensive. Therefore, a method of machine selection is devised based on tolerance range. In this study, mathematical formulation is developed for parallelism and perpendicularity constraints of a surface using the concept of torsors. Torsors represent any feature (plane, cylinder, point, etc.) only in terms of its degree of freedom. Torsors also have the advantage of being added during tolerance stack-up. Mathematical equations are developed for parallelism and perpendicularity as the function of surface parameters i.e. surface angles. Alternatively, the developed model can be used to control surface parameters to obtain the required tolerances. Maximum and minimum values of system parameters are calculated. Assuming that the parameters of the manufactured parts follow statistical distribution, random values of the surface parameters are generated and compared with the parallelism and perpendicularity constraints. It is concluded that the behavior of parallelism and perpendicularity tolerances is more precise if we control individual parameters. The results are in compliance with the proposed idea that a less precise machine can produce high quality product within a specific tolerance range as the machine need not control parallelism but only the dependent variables.

Processing multiple gap dependencies: Forewarned is forearmed

Abstract Many studies have shown that when forming a filler-gap dependency, comprehenders attempt to posit a gap site in advance of the input. However, it remains an open question what information they use to determine gap locations. The current study investigates parallelism in coordinate extraction structures, and asks whether comprehenders use parallelism constraints to structure their expectations about upcoming gap sites. Using a filled-gap paradigm, Experiments 1 and 2 show that comprehenders rely on parallelism to restrict the search for upcoming gap sites to specific locations in sentences with coordinate extraction. Experiment 3 shows that this effect cannot be reduced to processing factors, but may be due to a grammatically-based constraint on parallel extraction. Together, these results shed new light on the source and scope of active processing and parallelism effects in comprehension.

Accelerating Performance Inference over Closed Systems by Asymptotic Methods

Recent years have seen a rapid growth of interest in exploiting monitoring data collected from enterprise applications for automated management and performance analysis. In spite of this trend, even simple performance inference problems involving queueing theoretic formulas often incur computational bottlenecks, for example upon computing likelihoods in models of batch systems. Motivated by this issue, we revisit the solution of multiclass closed queueing networks, which are popular models used to describe batch and distributed applications with parallelism constraints. We first prove that the normalizing constant of the equilibrium state probabilities of a closed model can be reformulated exactly as a multidimensional integral over the unit simplex. This gives as a by-product novel explicit expressions for the multiclass normalizing constant. We then derive a method based on cubature rules to efficiently evaluate the proposed integral form in small and medium-sized models. For large models, we propose novel asymptotic expansions and Monte Carlo sampling methods to efficiently and accurately approximate normalizing constants and likelihoods. We illustrate the resulting accuracy gains in problems involving optimization-based inference.

Accelerating Performance Inference over Closed Systems by Asymptotic Methods

Recent years have seen a rapid growth of interest in exploiting monitoring data collected from enterprise applications for automated management and performance analysis. In spite of this trend, even simple performance inference problems involving queueing theoretic formulas often incur computational bottlenecks, for example upon computing likelihoods in models of batch systems. Motivated by this issue, we revisit the solution of multiclass closed queueing networks, which are popular models used to describe batch and distributed applications with parallelism constraints. We first prove that the normalizing constant of the equilibrium state probabilities of a closed model can be reformulated exactly as a multidimensional integral over the unit simplex. This gives as a by-product novel explicit expressions for the multiclass normalizing constant. We then derive a method based on cubature rules to efficiently evaluate the proposed integral form in small and medium-sized models. For large models, we propose novel asymptotic expansions and Monte Carlo sampling methods to efficiently and accurately approximate normalizing constants and likelihoods. We illustrate the resulting accuracy gains in problems involving optimization-based inference.

Eyeriss

Deep convolutional neural networks (CNNs) are widely used in modern AI systems for their superior accuracy but at the cost of high computational complexity. The complexity comes from the need to simultaneously process hundreds of filters and channels in the high-dimensional convolutions, which involve a significant amount of data movement. Although highly-parallel compute paradigms, such as SIMD/SIMT, effectively address the computation requirement to achieve high throughput, energy consumption still remains high as data movement can be more expensive than computation. Accordingly, finding a dataflow that supports parallel processing with minimal data movement cost is crucial to achieving energy-efficient CNN processing without compromising accuracy. In this paper, we present a novel dataflow, called row-stationary (RS), that minimizes data movement energy consumption on a spatial architecture. This is realized by exploiting local data reuse of filter weights and feature map pixels, i.e., activations, in the high-dimensional convolutions, and minimizing data movement of partial sum accumulations. Unlike dataflows used in existing designs, which only reduce certain types of data movement, the proposed RS dataflow can adapt to different CNN shape configurations and reduces all types of data movement through maximally utilizing the processing engine (PE) local storage, direct inter-PE communication and spatial parallelism. To evaluate the energy efficiency of the different dataflows, we propose an analysis framework that compares energy cost under the same hardware area and processing parallelism constraints. Experiments using the CNN configurations of AlexNet show that the proposed RS dataflow is more energy efficient than existing dataflows in both convolutional (1.4× to 2.5×) and fully-connected layers (at least 1.3× for batch size larger than 16). The RS dataflow has also been demonstrated on a fabricated chip, which verifies our energy analysis.

Parallelizing image feature extraction algorithms on multi-core platforms

Currently, multimedia data has become one of the most important data types processed and transferred over the Internet. To extract useful information from a huge amount of such data, SIFT and SURF, as two most popular image feature extraction algorithms, have been widely used in many applications running on multi-core platforms. However, limited parallelism in existing designs makes it hard or impossible to apply them in many applications with real-time requirements. Therefore, it has become one of the major challenges to improve the processing speed of image feature extraction algorithms.In this paper, we first analyze the parallelism constraints in the algorithms, such as imbalanced workloads and indeterminate time distributions. Based on such analyses, we present an adaptive pipeline parallel scheme (AD-PIPE) to adjust the thread number in different stages according to their workloads dynamically, which achieves a balanced partition for constant input workloads. Furthermore, we also implement a power efficient version (AE-PIPE) for AD-PIPE through scheduling threads based on variable input workloads. Experimental results show that AD-PIPE achieves a speedup of 16.88X and 20.33X respectively over SIFT and SURF on a 16-core machine. Moreover, AE-PIPE achieves up to 52.94% and 58.82% power saving with only 3% performance loss.

Subspace Learning of Dynamics on a Shape Manifold: A Generative Modeling Approach

In this paper, we propose a novel subspace learning algorithm of shape dynamics. Compared to the previous works, our method is invertible and better characterizes the nonlinear geometry of a shape manifold while retaining a good computational efficiency. In this paper, using a parallel moving frame on a shape manifold, each path of shape dynamics is uniquely represented in a subspace spanned by the moving frame, given an initial condition (the starting point and starting frame). Mathematically, such a representation may be formulated as solving a manifold-valued differential equation, which provides a generative modeling of high-dimensional shape dynamics in a lower dimensional subspace. Given the parallelism and a path on a shape manifold, the parallel moving frame along the path is uniquely determined up to the choice of the starting frame. With an initial frame, we minimize the reconstruction error from the subspace to shape manifold. Such an optimization characterizes well the Riemannian geometry of the manifold by imposing parallelism (equivalent as a Riemannian metric) constraints on the moving frame. The parallelism in this paper is defined by a Levi-Civita connection, which is consistent with the Riemannian metric of the shape manifold. In the experiments, the performance of the subspace learning is extensively evaluated using two scenarios: 1) how the high dimensional geometry is characterized in the subspace and 2) how the reconstruction compares with the original shape dynamics. The results demonstrate and validate the theoretical advantages of the proposed approach.

Balancing modular transfer lines with serial–parallel activation of spindle heads at stations

The paper deals with an as yet unexplored combinatorial optimization problem concerning balancing complex transfer lines in the machining/process environment. In contrast to similar problems for assembly lines, in transfer line balancing, tasks are grouped into blocks. All tasks of each block are executed simultaneously (in parallel) by one piece of equipment (spindle head). For the transfer lines considered in this paper, spindle heads at each station are activated in serial–parallel order. The set of all available spindle heads is known beforehand. Precedence, cycle time, compatibility, and parallelism constraints for the blocks and tasks are given. The line investment cost is estimated by the sum of block and station costs. The problem is to assign all tasks (using the available blocks) such that all constraints are respected and line investment cost is at a minimum. This paper focuses on solving the problem via a branch-and-bound algorithm. An approach for obtaining an efficient lower bound is offered, based on a reduction of the initial problem to a set partitioning problem. Computational experiments reveal that the proposed approach is efficient mathematically and can be used to solve practical transfer line design problems of a medium size.

Both parallelism and orthogonality are used to perceive 3D slant of rectangles from 2D images

A 2D perspective image of a slanted rectangular object is sufficient for a strong 3D percept. Two computational assumptions that could be used to interpret 3D from images of rectangles are as follows: (1) converging lines in an image are parallel in the world, and (2) skewed angles in an image are orthogonal in the world. For an accurate perspective image of a slanted rectangle, either constraint implies the same 3D interpretation. However, if an image is rescaled, the 3D interpretations based on parallelism and orthogonality generally conflict. We tested the roles of parallelism and orthogonality by measuring perceived depth within scaled perspective images. Stimuli were monocular images of squares, slanted about a horizontal axis, with an elliptical hole. Subjects judged the length-to-width ratio of the holes, which provided a measure of perceived depth along the object. The rotational alignment of squares within their surface plane was varied from 0 degrees (trapezoidal projected contours) to 20 degrees (skewed projected contours). In consistent-cue conditions, images were accurate projections of either a 10 degree- or 20 degree-wide square, with slants of 75 degrees and 62 degrees, respectively. In cue-conflict conditions, images were generated either by magnifying a 10 degrees image to have a projected size of 20 degrees or by minifying a 20 degree image to have a projected size of 10 degrees. For the aligned squares, which do not produce a conflicting skew cue, we found that subjects' judgments depended primarily on projected size and not on the size used to generate the prescaled images. This is consistent with reliance on the convergence cue, corresponding to a parallelism assumption. As squares were rotated away from alignment, producing skewed projected contours, judgments were increasingly determined by the original image size. This is consistent with use of the skew cue, corresponding to an orthogonality assumption. Our results demonstrate that both parallelism and orthogonality constraints are used to perceive depth from linear perspective.

A Note on ‘‘Cyclic Linearization’’

‘‘Cyclic Linearization’’ (CL) is similar in its intent to my own ‘‘Representation Theory’’ (Williams 1998, 2003) (RT) in that it seeks to explain some parallelism constraints in terms of general architecture, and in fact both were inspired by Holmberg’s generalization and the general properties of Object Shift in the formulation of central tenets. But here, I will explore a difference in the models that has more to do with WH movement.

Mapping Interleaving Laws to Parallel Turbo Decoder Architectures

For high data rate applications, the implementation of iterative turbo-like decoders requires the use of parallel architectures posing some collision-free constraints to the reading/writing process in the soft-input soft-output (SISO) decoders. Contrary to the literature belief, we prove in this paper that the parallelism constraints can be met by any permutation law employed by the turbo-interleaver, and we give a constructive method to satisfy those constraints.

Automatic formal verification for scheduled VLIW code

VLIW processors are attractive for many embedded applications, but VLIW code scheduling, whether by hand or by compiler, is extremely challenging. In this paper, we extend previous work on automated verification of low-level software to handle the complexity of modern, aggressive VLIW designs, e.g., the exposed parallelism, pipelining, and resource constraints. We implement these ideas into a prototype tool for verifying short sequences of assembly code for TI's C62x family of VLIW DSPs, and demonstrate the effectiveness of the tool in quickly verifying, or finding bugs in, two difficult-to-analyze code segments.

Multiprocessor Scheduling with Machine Allotment and Parallelism Constraints

Abstract. Modern computer systems distribute computation among several machines to speed up the execution of programs. Yet, setup and communication costs, as well as parallelism constraints, bound the number of machines that can share the execution of a given application, and the number of machines by which it can be processed simultaneously . We study the resulting scheduling problem, stated as follows. Given a set of n jobs and m uniform machines, assign the jobs to the machines subject to parallelism and machine allotment constraints, such that the overall completion time of the schedule (or makespan ) is minimized. Indeed, the multiprocessor scheduling problem (where each job can be processed by a single machine) is a special case of our problem; thus, our problem is strongly NP-hard. We present a (1+ α) -approximation algorithm for this problem, where α ∈ (0,1] depends on the minimal number of machine allotments and the minimal parallelism allowed for any job. Also, we show that when the maximal number of machines that can share the execution of a job is some fixed constant, our problem has a polynomial time approximation scheme ; for other special cases we give optimal polynomial time algorithms. Finally, through the relation of our problem to the classic preemptive scheduling problem on multiple machines, we shed some fresh light on what is known in scheduling folklore as the power of preemption.

Interactive optimization of 3D shape and 2D correspondence using multiple geometric constraints via POCS

The traditional approach of handling motion tracking and structure from motion (SFM) independently in successive steps exhibits inherent limitations in terms of achievable precision and incorporation of prior geometric constraints about the scene. This paper proposes a projections onto convex sets (POCS) framework for iterative refinement of the measurement matrix in the well-known factorization method to incorporate multiple geometric constraints about the scene, thereby improving the accuracy of both 2D feature point tracking and 3D structure estimates. Regularities in the scene, such as points on line and plane and parallel lines and planes, that can be interactively identified and marked at each POCS iteration, enforce rank and parallelism constraints on appropriately defined local measurement matrices, one for each constraint. The POCS framework allows for the integration of the information in each of these local measurement matrices into a single measurement matrix that is "closest" to the initial observed measurement matrix in Frobenius norm, which is then factored in the usual manner. Experimental results demonstrate that the proposed interactive POCS framework consistently improves both 2D correspondences and 3D shape/motion estimates and similar results cannot be achieved by enforcing these constraints as either post or preprocessing.

Les conjonctions doubles

We present a syntactic analysis of the French “double conjunctions” class, which contains ninety one items: some of them are subordinating items, and in majority coordinating items. We analyse their specific syntactic properties by means of the “parallelism constraints” (see Z. S. Harris (1968)), related to the specific nature and order of these items and their sentences. Study of restrictions in their positions and combinations reveals complete independancy between “double conjunctions” and some homonymous “simple” ones, excluding any analyse of the last ones from the first ones.