THE UNCONDITIONAL CONVERGENT DIFFERENCE METHODS WITH INTRINSIC PARALLELISM FOR QUASILINEAR PARABOLIC SYSTEMS WITH TWO DIMENSIONS

In the present work we are going to solve the boundary value problem for the quasilinear parabolic systems of partial differential equations with two space dimensions by the finite difference method with intrinsic parallelism.Some fundamental behaviors of general finite difference schemes with intrinsic parallelism for the mentioned problems are studied.By the method of a priori estimation of the discrete solutions of the nonlinear difference systems,and the interpolation formulas of the various norms of the discrete functions and the fixed-point technique in finite dimensional Euclidean space,the existennce of the discrete vector solutions of the nonliear difference system with intrinsic parallelism are proved .Moreover the convergence of the discrete vector solutions of these difference schemes to the unique generalizd solution of the original quasilinear parabolic problem is proved.     

Experimenting with parallelism for the instantiation of ASP programs

In the last few years, microprocessor technologies have been moving towards multi-core architectures, in order to improve performance as well as reduce power consumption. This makes real Symmetric MultiProcessing (SMP) available even on non-dedicated machines, and paves the way to the development of better performing software. Notably, the recent application of Answer Set Programming (ASP) in different emerging areas, such as knowledge management or information extraction/integration, shows that performance is a crucial issue also for ASP systems. Among the tasks performed by such systems, the instantiation process, which consists of generating a variable-free program equivalent to the input one, is one of the most expensive from a computational viewpoint, especially in the case of huge input data. In this paper a new strategy exploiting parallelism for the instantiation of ASP programs is proposed. An implementation of this strategy and its integration with the grounding module of the DLV system is discussed. The results of an experimental analysis are also presented, which confirm that the strategy is effective in making ASP instantiation more efficient.     

Parallelism error characterization with mechanical and interferometric methods

Two methods for measuring the parallelism error of gauge blocks are discussed. The first method makes use of contact probes, sampling the gauge thickness at fixed locations; the parallelism error is given by the difference between maximum and minimum thickness over the sampled area. The second method is based on interferometry, producing the height maps of the opposing surfaces with respect to one another; the parallelism error is obtained after processing of the interferometric data. The two methods have been implemented in different laboratories, and compared by measuring five different gauge blocks. The results are reported, showing that the first method is more straightforward and simple, but requires prior calibration and provides information only on a limited number of sample points. The second method is more demanding in terms of equipment and measuring procedure, but in principle it is an absolute method, and provides information over a large number of data points.     

A note on parallel and alternating time

A long standing open question in complexity theory over the reals is the relationship between parallel time and quantifier alternation. It is known that alternating digital quantifiers is weaker than parallel time, which in turn is weaker than alternating unrestricted (real) quantifiers. In this note we consider some complexity classes defined through alternation of mixed digital and unrestricted quantifiers in different patterns. We show that the class of sets decided in parallel polynomial time is sandwiched between two such classes for different patterns.     

Using diversification,communication and parallelism to solve mixed-integer linear programs

Performance variability of modern mixed-integer programming solvers and possible ways of exploiting this phenomenon present an interesting opportunity in the development of algorithms to solve mixed-integer linear programs (MILPs). We propose a framework using multiple branch-and-bound trees to solve MILPs while allowing them to share information in a parallel execution. We present computational results on instances from MIPLIB 2010 illustrating the benefits of this framework.     

Two-step-by-two-step PIRK-type PC methods based on Gauss-Legendre collocation points

This paper concerns with parallel predictor-corrector (PC) iteration methods for solving nonstiff initial-value problems (IVPs) for systems of first-order differential equations. The predictor methods are based on Adams-type formulas. The corrector methods are constructed by using coefficients of s-stage collocation Gauss-Legendre Runge-Kutta (RK) methods based on c₁,…,c_s and the 2s-stage collocation RK methods based on c₁,…,c_s,1+c₁,…,1+c_s. At nth integration step, the stage values of the 2s-stage collocation RK methods evaluated at t_n+(1+c₁)h,…,t_n+(1+c_s)h can be used as the stage values of the collocation Gauss-Legendre RK method for (n+2)th integration step. By this way, we obtain the corrector methods in which the integration processes can be proceeded two-step-by-two-step. The resulting parallel PC iteration methods which are called two-step-by-two-step (TBT) parallel-iterated RK-type (PIRK-type) PC methods based on Gauss-Legendre collocation points (two-step-by-two-step PIRKG methods or TBTPIRKG methods) give us a faster integration process. Fixed step size applications of these TBTPIRKG methods to the three widely used test problems reveal that the new parallel PC iteration methods are much more efficient when compared with the well-known parallel-iterated RK methods (PIRK methods) and sequential codes ODEX, DOPRI5 and DOP853 available from the literature.     

飞秒激光制造微型光纤法布里-珀罗干涉传感器

基于飞秒激光光刻技术,提出了光纤端面光刻制造微型光纤法布里-珀罗干涉传感器的方案.该方案解决了传统飞秒激光制造光纤F-P传感器的平行度差的问题,制造了对比度超过20 dB的高灵敏度和高分辨率的微型F-P光纤传感器件.该制造方法简单、参量可控,制作的器件可应用恶劣温度条件下应变的精确测量.     

THE UNCONDITIONAL STABLE DIFFERENCE METHODS WITH INTRINSIC PARALLELISM FOR TWO DIMENSIONAL SEMILINEAR PARABOLIC SYSTEMS

In this paper we are going to discuss the difference schemes with intrinsic parallelism for the boundary value problem of the two dimesional semilinear parabolic systems.The unconditional stability of the general finite difference schemes with intrinsic parallelism is justified in the sense of the continuous dependence of the discrete vector solution of the difference schemes on the discrete data of the original problems in the discrete W2^(2,1) norms.Then the uniqueness of the discrete vector solution of this difference scheme follows as the consequence of the stability.     

利用循环展开最大化软件流水线性能

软件流水线通过重叠连续的循环实体来实现有效的精细调度．然而,其性能可能受限制于循环里缺乏足够的并行操作或者资源需求．‘‘先展开后调度”技术在进行软件流水线调度之前先展开循环,从而能够发现更多的并行操作和充分利用关键资源．研究循环展开如何影响软件流水线的性能和资源利用,并进一步提出如何选择优化的循环展开次数．     

Parallel iteration across the steps of high-order Runge-Kutta methods for nonstiff initial value problems

For the parallel integration of nonstiff initial value problems (IVPs), three main approaches can be distinguished: approaches based on “parallelism across the problem”, on “parallelism across the method” and on “parallelism across the steps”. The first type of parallelism does not require special integration methods and can be exploited within any available IVP solver. The method-parallelism approach received much attention, particularly within the class of explicit Runge-Kutta methods originating from fixed point iteration of implicit Runge-Kutta methods of Gaussian type. The construction and implementation on a parallel machine of such methods is extremely simple. Since the computational work per processor is modest with respect to the number of data to be exchanged between the various processors, this type of parallelism is most suitable for shared memory systems. The required number of processors is roughly half the order of the generating Runge-Kutta method and the speed-up with respect to a good sequential IVP solver is about a factor 2. The third type of parallelism (step-parallelism) can be achieved in any IVP solver based on predictor-corrector iteration and requires the processors to communicate after each full iteration. If the iterations have sufficient computational volume, then the step-parallel approach may be suitable for implementation on distributed memory systems. Most step-parallel methods proposed so far employ a large number of processors, but lack the property of robustness, due to a poor convergence behaviour in the iteration process. Hence, the effective speed-up is rather poor. The dynamic step-parallel iteration process proposed in the present paper is less massively parallel, but turns out to be sufficiently robust to achieve speed-up factors up to 15.