Algorithm and Data Structures for Xeon Phi

Simulation of particle transport is critical for a great many of scientific and engineering domains. The Monte Carlo (MC) method is one of the most important numerical methods for the simulation of particle transport, and can simulate many complex types of particle transport. But the computation requirement of the MC simulation is very large. In 2010, Intel announced the Intel Many Integrated Core (MIC) architecture, which consists of many simple general-purpose cores and supports the well-known shared-memory execution model that is the base of most nodes in HPC machines. On account of the independence of simulation of each particle in the MC method, it is well-suited to accelerate the MC simulation on MIC. In this paper, an algorithm named MCNP-MIC based MIC is presented for MC simulation of neutron transport in the context of deep penetration problem, which includes the development of parallel random generator, the assignment of particle number based thread number and the design of high efficiency data structures for parallelism. Eventually, we get the results as follows: with the same problem scale and computational accuracy, the MCNPMIC algorithm has achieved roughly 5.6-fold speedup running on a 57-core MIC chip in comparison with the serial MCNP algorithm on an Intel Xeon E5-2670 CPU.

Intel Many Integrated Core (MIC) ushers in a new era of supercomputing speed, performance, and compatibility. It allows the developers to run code at trillions of calculations per second using the familiar programming model. In this paper, we present our results of optimizing the updated Goddard shortwave radiation Weather Research and Forecasting (WRF) scheme on Intel Many Integrated Core Architecture (MIC) hardware. The Intel Xeon Phi coprocessor is the first product based on Intel MIC architecture, and it consists of up to 61 cores connected by a high performance on-die bidirectional interconnect. The co-processor supports all important Intel development tools. Thus, the development environment is familiar one to a vast number of CPU developers. Although, getting a maximum performance out of Xeon Phi will require using some novel optimization techniques. Those optimization techniques are discusses in this paper. The results show that the optimizations improved performance of the original code on Xeon Phi 7120P by a factor of 1.3x.

Intel Xeon Phi is a new addition to the family of powerful parallel accelerators. The range of its potential applications in computationally driven research is broad; however, at present, the repository of scientific codes is still relatively limited. In this study, we describe the development and benchmarking of a parallel version of eFindSite, a structural bioinformatics algorithm for the prediction of ligand-binding sites in proteins. Implemented for the Intel Xeon Phi platform, the parallelization of the structure alignment portion of eFindSite using pragma-based OpenMP brings about the desired performance improvements, which scale well with the number of computing cores. Compared to a serial version, the parallel code runs 11.8 and 10.1 times faster on the CPU and the coprocessor, respectively; when both resources are utilized simultaneously, the speedup is 17.6. For example, ligand-binding predictions for 501 benchmarking proteins are completed in 2.1 hours on a single Stampede node equipped with the Intel Xeon Phi card compared to 3.1 hours without the accelerator and 36.8 hours required by a serial version. In addition to the satisfactory parallel performance, porting existing scientific codes to the Intel Xeon Phi architecture is relatively straightforward with a short development time due to the support of common parallel programming models by the coprocessor. The parallel version of eFindSite is freely available to the academic community at www.brylinski.org/efindsite.

Modern heterogeneous high-end computing systems featuring many-core co-processors/accelerators pose tough challenges to parallelize and optimize real-world scientific codes. In this paper, we demonstrate highly scalable 3D Lattice Boltzmann multiphase flow simulations on the heterogeneous Tianhe-2 su-percomputer using MPI+OpenMP+SIMD. We highlight the use of the OpenMP4.5 accelerator programming model to collaborate CPUs and Intel Many Integrated Cores (MIC) co-processors, and present a range of optimizations to exploit hierarchical parallelism on the heterogeneous architecture. With SIMD-friendly data structures, computation reordering, cache blocking and auto-vectorization, we dramatically optimize the single-thread performance for typical LBM computational kernels on a CPU and MIC core. After implementation of shared-memory OpenMP threading, we further improve the OpenMP performance by enabling huge paging and OpenMP collapse clause on the thread-rich MIC architecture. To enhance the collaborative efficiency among intra-node CPUs and co-processors, we propose a flexible load balance model with heterogeneous domain decomposition for CPU-MIC task allocation, as well as asynchronous offloading to overlap operations of CPUs and multiple MICs. We effectively overlap all levels of CPU-MIC computation/communication and minimize the cost of halo exchanges as far as possible in large-scale scenario. 3D liquid and gases multi-phase cases simulating drop impact with gravity effect using D3Q19 Lattice Boltzmann discretization and Shan-Chen BGK collision model are presented, achieving a weak parallel efficiency of above 80% in going from 128 to 2048 compute nodes.

This paper outlines the parallelisation and vectorisation methods we have used to port a LU decomposition library to the Xeon Phi co-processor. We ported a LU factorisation algorithm, which utilizes the Gaussian elimination method to perform the decomposition, using Intel LEO directives, OpenMP 4.0 directives, Intel's Cilk array notation, and vectorisation directives. We compare the performance achieved with these different methods, investigate the cost of data transfer on the overall time to solution, and analyse the impact of these optimization and parallelisation techniques on code running on the host processors as well. The results show that performance can be improved on the Xeon Phi by optimising the memory operations, and that Cilk array notation can benefit this benchmark on standard processors but do not have the same impact on the Xeon Phi co-processor. We have also demonstrated cases where the Xeon Phi will compute our implementations faster than we can run them on a node of a HPC system, and that our implementations are not as efficient as the LU factorisation implemented in the mkl library.

Summary form only given, as follows. The Common Information Model (CIM) is a common standardized data structure, developed by EPRI, similar to most Energy Management Systems (EMS) proprietary data structures. There have been developed filters that will map most of the major proprietary EMS data systems to the CIM data structures. The CIM is an excellent vehicle, for example, to run EMS applications and to merge data from multiple EMS systems. By using the CIM applications can run on a variety of EMS systems without changing the data structure. The panelist will show the CIM tools, Application Programming Interfaces (API) and a variety of uses and application of the CIM. A record of the panel discussion was not made available for publication as part of the conference proceedings.

Combining methods are highly effective for implementing concurrent queues and stacks. These data structures induce a heavy competition on one or two contention points. However, it was not known whether combining methods could be made effective for parallel scalable data structures that do not have a small number of contention points. In this paper, we introduce local combining on-demand, a new combining method for highly parallel data structures. The main idea is to apply combining locally for resources on which threads contend. We demonstrate the use of local combining on-demand on the common linked-list data structure. Measurements show that the obtained linked-list induces a low overhead when contention is low and outperforms other known implementations by up to 40% when contention is high.KeywordsConcurrent Data-StructuresMultiprocessorsSynchronization

Accelerators are used in about 13% of the current Top500 List. Supercomputers leveraging accelerators grew by a factor of 2.2x in 2012 and are expected to completely dominate the Top500 by 2015. Though most of these deployments use NVIDIA GPGPU accelerators, Intel's Xeon Phi architecture will likely grow in popularity in the coming years. Unfortunately, there are few studies analyzing the performance and energy efficiency of systems leveraging the Intel Xeon Phi. We extend our systemic measurement methodology to isolate system power by component including accelerators. We use this methodology to present a detailed study of the performance-energy tradeoffs of the Xeon Phi architecture. We demonstrate the portability of our approach by comparing our Xeon Phi results to the Intel multicore Sandy Bridge host processor and the NVIDIA Tesla GPU for a wide range of HPC applications. Our results help explain limitations in the power-performance scalability of HPC applications on the current Intel Xeon Phi architecture.

This paper presents an optimal proposed allocating procedure for hybrid wind energy combined with proton exchange membrane fuel cell (WE/PEMFC) system to improve the operation performance of the electrical distribution system (EDS). Egypt has an excellent wind regime with wind speeds of about 10 m/s at many areas. The disadvantage of wind energy is its seasonal variations. So, if wind power is to supply a significant portion of the demand, either backup power or electrical energy storage (EES) system is needed to ensure that loads will be supplied in reliable way. So, the hybrid WE/PEMFC system is designed to completely supply a part of the Egyptian distribution system, in attempt to isolate it from the grid. However, the optimal allocation of the hybrid units is obtained, in order to enhance their benefits in the distribution networks. The critical buses that are necessary to install the hybrid WE/ PEMFC system, are chosen using sensitivity analysis. Then, the binary Crow search algorithm (BCSA), discrete Jaya algorithm (DJA) and binary particle swarm optimization (BPSO) techniques are proposed to determine the optimal operation of power systems using single and multi-objective functions (SOF/MOF). Then, the results of the three optimization techniques are compared with each other. Three sensitivity factors are employed in this paper, which are voltage sensitivity factor (VSF), active losses sensitivity factor (ALSF) and reactive losses sensitivity factor (RLSF). The effects of the sensitivity factors (SFs) on the SOF/MOF are studied. The improvement of voltage profile and minimizing active and reactive power losses of the EDS are considered as objective functions. Backward/forward sweep (BFS) method is used for the load flow calculations. The system load demand is predicted up to year 2022 for Mersi-Matrouh City as a part of Egyptian distribution network, and the design of the hybrid WE/PEMFC system is applied. The PEMFC system is designed considering simplified mathematical expressions. The economics of operation of both WE and PEMFC system are also presented. The results prove the capability of the proposed procedure to find the optimal allocation for the hybrid WE/PEMFC system to improve the system voltage profile and to minimize both active and reactive power losses for the EDS of Mersi-Matrough City.

With rapid advancement in technologies used in solid-state electronics, nowadays, the conversion and control of electrical power has become easy due to inclusion of power electronics-based devices. With the help of these devices, accurate and smooth control of electrical power conversion can be consummated in three-phase voltage source inverters. But uses of power electronics devices also result in Total Harmonic Distortion (THD), which distorts overall system performance. Eliminating some selected harmonic in a system can quickly be done by using optimization techniques. This manuscript focuses on analysis of harmonics in a 3-phase voltage source inverter using Gravitational Search Algorithm (GSA) and Particle Swarm Optimization (PSO) technique.

Current research in the field of computational biology often involves simulations on high-performance computer clusters. It is crucial that the code of such simulations is efficient and correctly reflects the model specifications.In this paper, we present an optimization strategy for agent-based simulations of biological dynamics using Intel Xeon Phi coprocessors, demonstrated by a prize-winning entry of the “Intel Modern Code Developer Challenge” competition. These optimizations allow simulating various biological mechanisms, in particular the simulation of millions of cells, their proliferation, movements and interactions in 3D space. Overall, our results demonstrate a powerful approach to implement and conduct very detailed and large-scale computational simulations for biological research. We also highlight the main difficulties faced when developing such optimizations, in particular the assessment of the simulation accuracy, the dependencies between different optimization techniques and counter-intuitive effects in the speed of the optimized solution. The overall speedup of 595 × shows a good parallel scalability.

The power balance is considered as most remarkable issue in power generation system. In this proposed work thermal power systems are connected through tie-line. Thermal power system is designed by considering reheat turbine with single stage, governor, and speed regulator unit. Proportional-Integral-Derivative (PID) controller is implemented to regulate the system operation. In this proposed work, the automatic generation control of three-area interconnected reheat thermal power generating system designed and discussed. The PID controller gain values are tuned with the help of more powerful evolutionary Bat Algorithm (BA) procedure. The proposed Bat algorithm tuned controller response was examined by comparing its performance with other optimization technique, namely Genetic Algorithm (GA) and Particle Swam Optimization (PSO) technique tuned controller response. Additionally, different cost functions-based bat algorithm optimized controller responses are presented and compared. The time domain specification parameters are considered for verifying the better-cost function for designing of controller. The simulated responses evident that proposed bat algorithm tuned controller output yield superior performance over GA & PSO tuned controller. Integral Time Square Error (ITSE) cost function-based BA tuned controller give better controlled response during sudden load demand condition in interconnected power system.

OpenACC is gaining momentum as an implicit and portable interface in porting legacy CPU-based applications to heterogeneous, highly parallel computational environment involving many-core accelerators such as GPUs and Intel Xeon Phi. OpenACC provides a set of loop directives similar to OpenMP for the parallelization and also to manage data movement, attaining functional portability across different heterogeneous devices; however, the performance portability of OpenACC is said to be insufficient due to the characteristics of different target devices, especially those regarding memory layouts, as automated attempts by the compilers to adapt is currently difficult. We are currently working to propose a set of directives to allow compilers to have better semantic information for adaptation; here, we particularly focus on data layout such as Structure of Arrays, advantageous data structure for GPUs, as opposed to Array of Structures, which exhibits good performance on CPUs. We propose a directive extension to OpenACC that allows the users to flexibility specify optimal layouts, even if the data structures are nested. Performance results show that we gain as much as 96 % in performance for CPUs and 165% for GPUs compared to programs without such directives, essentially attaining both functional and performance portability in OpenACC.

Intel Xeon Phi acceleration of Hybrid Total FETI solver

This article presents parallel implementations of the Mass Approximation Distance Algorithm for self-gravity calculation on Xeon Phi. The proposed method is relevant for performing simulations on realistic systems modeling small astronomical bodies, which are agglomerates of thousand/million of particles. Specific strategies and optimizations are described for execution on the Xeon Phi architecture. The experimental analysis evaluates the computational efficiency of the proposed implementations on realistic scenarios, reporting the best options for the implementation. Specific performance improvements of up to 146.4\(\times \) are reported for scenarios with more than one million particles.