Interprocessor Communication Research Articles

High Performance Computing Framework for Tera-Scale Database Search of Mass Spectrometry Data.

Database peptide search algorithms deduce peptides from mass spectrometry (MS) data. There has been substantial effort in improving their computational efficiency to achieve larger and more complex systems biology studies. However, modern serial and high-performance computing (HPC) algorithms exhibit sub-optimal performance mainly due to their ineffective parallel designs (low resource utilization), and high overhead costs. We present an HPC framework, called HiCOPS, for efficient acceleration of the database peptide search algorithms on distributed-memory supercomputers. HiCOPS provides, on average, more than 10-fold improvement in speed, and superior parallel performance over several existing HPC database search software. We also formulate a mathematical model for performance analysis and optimization, and report near-optimal results for several key metrics including strong-scale efficiency, hardware utilization, load-balance, inter-process communication and I/O overheads. The core parallel design, techniques, and optimizations presented in HiCOPS are search-algorithm independent and can be extended to efficiently accelerate the existing and future algorithms and software.

PARALLEL IMPLEMENTATION OF THE YANENKO METHOD FOR SOLVING THE HEAT EQUATION

In this article, we will consider the parallel implementation of the Yanenko algorithm for the two-dimensional heat equation, and the sweep method was used to numerically solve the heat equation. The implementation of the sequential program is carried out simply in two-part steps by the longitudinal-transverse run, however, parallelization of two fractional steps with an indefinite scheme is difficult due to the creation of inter-process communication of data. In the course of the study, a parallel data distribution with one-dimensional decompositions is shown in the application of the Yanenko method for calculating heat conductivity. The results of parallelization of this task using the 1D decomposition were obtained and acceleration and efficiency images were analyzed in order to evaluate the parallel program. Currently, modeling of processes by numerical solution of differential equations is widely used in every field of Science, the most common methods bring the differential problem to a system of linear algebraic equations, methods that solve such systems include various startup options. The emergence and development of computing systems using Multi-Core processors and graphics accelerators make the problem of startup parallelization relevant; the results of the study are used for teaching in research institutes and universities.

Evaluating crop-soil-water dynamics in waterlogged areas using a coupled groundwater-agronomic model

Waterlogging on croplands has been a known problem for a long time, leading to adverse social, physical, economic and environmental issues. To better solve the problem, the complicated plant-soil-water dynamics system needs to be better understood. The challenge is to simulate the interactions between the components in the systems. There are models that simulate plant-soil-water system but either run the processes independently leading to inaccuracy or has high invasiveness of using integrated models. This paper presents a tightly coupled model, DayCent-MODFLOW, that links a 3D ground-water flow (MODFLOW) model and a 1D agroecosystem model (DayCent). DayCent is responsible for plant-soil-water dynamic in the root zone, whereas MODFLOW simulates head and groundwater flow in the saturated zone of the aquifer. DayCent passes deep percolation from the soil profile to the water table and, under conditions of waterlogging in which the water table is within the soil profile, DayCent soil hydrologic processes are constrained by the presence of the water table simulated by MODFLOW. The coupling is achieved by adopting a parallel inter-process communication technique MPI (Message Passing Interface). The model is applied to a waterlogged agricultural area (22 km2) in northern Colorado, USA and tested against groundwater head and rates of evapotranspiration (ET). The model runs in parallel with multiple processes on the largest AWS Linux server. Groundwater heads match measured heads to a reasonable degree, and ET rates match reference ET and are highly correlated with crop type. Results show the strong hydrologic interaction between the two models. Greenhouse gas emissions from soil (N2O and CH4) were also estimated by the model under the waterlogged conditions. Although the model can be used to simulate any plant-soil-aquifer system, no matter the depth of the water table, results from this study show that the model can be used to assess crop productivity, recharge, ET, and greenhouse gas emissions in areas of shallow groundwater.

An efficient and scalable parallel mapping of pulse-Doppler radar signal processing chain on a multi-core DSP

Digital Pulse-Doppler radar chain consists of signal processing algorithms that require high computing power. Multi-processor and multi-core parallel embedded machines are one of the solutions to meet real-time constraints of many radar applications. In this paper, we proposed efficient and scalable parallelization methods of the Pulse-Doppler radar signal processing chain. First, we evaluated Open Multi Processing (OpenMP) to identify its best scheduling technique in order to exploit efficiently the available computing cores. Then, we have proposed new parallel and scalable approaches based on direct memory access (DMA) and inter-processor communication (IPC) techniques, combined with the best OpenMP scheduling method to accelerate radar signal processing chain. To prove the scalability of our proposed parallel approaches, two radar use cases with different real-time and memory constraints have been experienced. We used the eight cores C6678 digital signal processor (DSP) as a target for all our implementations. The obtained results show an overall parallel efficiency of 95%, which is better than the best state-of-the-art implementations.

Online data analysis and reduction: An important Co-design motif for extreme-scale computers

A growing disparity between supercomputer computation speeds and I/O rates means that it is rapidly becoming infeasible to analyze supercomputer application output only after that output has been written to a file system. Instead, data-generating applications must run concurrently with data reduction and/or analysis operations, with which they exchange information via high-speed methods such as interprocess communications. The resulting parallel computing motif, online data analysis and reduction (ODAR), has important implications for both application and HPC systems design. Here we introduce the ODAR motif and its co-design concerns, describe a co-design process for identifying and addressing those concerns, present tools that assist in the co-design process, and present case studies to illustrate the use of the process and tools in practical settings.

A parallel optimization method for kernel correlation filter based on multi-core DSP

Target tracking is an important branch in the field of computer vision. In recent years, parallel computing efficiency becomes the key to improving real-time performance in multi-core DSP platforms. This paper proposes the concept of execution parallelism based on Open MP, optimizes Kernel Correlation Filter (KCF) with execution parallelism parameters, and compares it with inter-processor Communication (IPC). This paper aims at the KCF algorithm on the TMS320C6678 platform for the multi-core parallel development of the above two methods. With multiple data-sets, the results are all higher than 60 frames/s. With physical verification by connecting the video board, parallel scheme based on Open MP can achieve a tracking effect of 75 FPS, which has certain engineering practical significance.

Identifying Degree and Sources of Non-Determinism in MPI Applications Via Graph Kernels

As the scientific community prepares to deploy an increasingly complex and diverse set of applications on exascale platforms, the need to assess reproducibility of simulations and identify the root causes of reproducibility failures increases correspondingly. One of the greatest challenges facing reproducibility issues at exascale is the inherent non-determinism at the level of inter-process communication. The use of non-deterministic communication constructs is necessary to boost performance, but communication non-determinism can also hamper software correctness and result reproducibility. To address this challenge, we propose a software framework for identifying the percentage and sources of communication non-determinism. We model parallel executions as directed graphs and leverage graph kernels to characterize run-to-run variations in inter-process communication. We demonstrate the effectiveness of graph kernel similarity as a proxy for non-determinism, by showing that these kernels can quantify the type and degree of non-determinism present in communication patterns. To demonstrate our framework's ability to link and quantify runtime non-determinism to root sources, demonstrate with present for an adaptive mesh refinement application, where our framework automatically quantifies the impact of function calls on non-determinism, and a Monte Carlo application, where our framework automatically quantifies the impact of parameter configurations on non-determinism.

An Inverse Kinematics Demonstration of a Custom Robot using C# and CoppeliaSim

Purpose: Inverse Kinematics (I.K.) is not as easy as Forward kinematics (F.K.), where we get a definite result. I.K. algorithm provides several possible solutions. From those finding the best solution is such a critical task. For standard robots which are commercially available in the market, the user is not concerned about I.K.'s complexity. They provide the control board and programming IDE to make it easy. However, when we develop a robotic arm from our D.H. parameter and driver board, complexity arises due to lots of difficulties for executing and successful completion. To make life easy, keeping CoppeliaSim background can eliminate the calculation overhead and get good results. The custom robot is running with less computation power. It may be a good approach. We are using C# for User Interaction. Following step by step, anyone can create a robust I.K. engine with little effort. The complete code is available in GitHub to test and experiment further. Design/Methodology/Approach: The data are propagated through Interprocess communication. For the user interaction, we use visual studio IDE using the most accessible language, C#. The user interaction data are sent to another application, CoppeliaSim, which calculates inverse kinematics, and effective results are displayed through robotic arm movement. Findings/Result: Implementing this procedure can get the excellent result of the robotics arm. Furthermore, by imposing the Value on the real robot, we can get effective results. It minimizes the research overhead on I.K. calculation. Originality/Value: Without knowing I.K. calculation complexity, receiving the Value, we can apply it to the real robot. Two issues we can solve here. One is the calculation, and another one is experiment overhead. Paper Type: Simulation-based Research.

An Inverse Kinematics Demonstration of a Custom Robot using C# and CoppeliaSim

Purpose: Inverse Kinematics (I.K.) is not as easy as Forward kinematics (F.K.), where we get a definite result. I.K. algorithm provides several possible solutions. From those finding the best solution is such a critical task. For standard robots which are commercially available in the market, the user is not concerned about I.K.'s complexity. They provide the control board and programming IDE to make it easy. However, when we develop a robotic arm from our D.H. parameter and driver board, complexity arises due to lots of difficulties for executing and successful completion. To make life easy, keeping CoppeliaSim background can eliminate the calculation overhead and get good results. The custom robot is running with less computation power. It may be a good approach. We are using C# for User Interaction. Following step by step, anyone can create a robust I.K. engine with little effort. The complete code is available in GitHub to test and experiment further. Design/Methodology/Approach: The data are propagated through Interprocess communication. For the user interaction, we use visual studio IDE using the most accessible language, C#. The user interaction data are sent to another application, CoppeliaSim, which calculates inverse kinematics, and effective results are displayed through robotic arm movement. Findings/Result: Implementing this procedure can get the excellent result of the robotics arm. Furthermore, by imposing the Value on the real robot, we can get effective results. It minimizes the research overhead on I.K. calculation. Originality/Value: Without knowing I.K. calculation complexity, receiving the Value, we can apply it to the real robot. Two issues we can solve here. One is the calculation, and another one is experiment overhead. Paper Type: Simulation-based Research.

GFCCLib: Scalable and efficient coupled-cluster Green's function library for accurately tackling many-body electronic structure problems

Coupled-cluster Green's function (GFCC) calculation has drawn much attention in the recent years for targeting the molecular and material electronic structure problems from a many-body perspective in a systematically improvable way. However, GFCC calculations on scientific computing clusters usually suffer from expensive higher dimensional tensor contractions in the complex space, expensive inter-process communication, and severe load imbalance, which limits it's use for tackling electronic structure problems. Here we present a numerical library prototype that is specifically designed for large-scale GFCC calculations. The design of the library is focused on a systematically optimal computing strategy to improve its scalability and efficiency. The performance of the library is demonstrated by the relevant profiling analysis of running GFCC calculations on remote giant computing clusters. The capability of the library is highlighted by computing a wide near valence band of a fullerene C60 molecule for the first time at the GFCCSD level that shows excellent agreement with the experimental spectrum. Program summaryProgram Title: GFCCLibCPC Library link to program files:https://doi.org/10.17632/j594wydctd.1Code Ocean capsule:https://doi.org/10.24433/CO.5131827.v1Licensing provisions: MIT LicenseProgramming language: C++Nature of problem: The applications of coupled cluster Green's function on large scale molecular electronic structure problems suffer from expensive higher dimensional tensor contractions in the complex space, expensive inter-process communication, and severe load imbalance. Tackling these issues are a key step in building high-performance coupled cluster Green's function library for its routine use in large scale molecular science.Solution method: We have developed a C++ library for large scale molecular GFCC calculations on high-performance computing clusters. We provide implementations for high dimensional tensor algebra for many-body methods (TAMM), Cholesky decomposition of high dimensional electron repulsion integral tensors, process group technique for mitigating load imbalance. The library is written in C++. The source code, tutorials and documentation are provided online. A continuous integration mechanism is set up to automatically run a series of regression tests and check code coverage when the codebase is updated.

A ROS-Based Energy Management System for a Prototype Fuel Cell Hybrid Vehicle

The automotive industry has been rapidly transforming and moving further from internal combustion engines, towards hybrid or electric vehicles. A key component for the successful adoption of the aforementioned approach is their Energy Management Systems (EMSs). In the proposed work, we describe in detail a custom EMS, with unique characteristics, which was developed and installed in a hydrogen-powered prototype vehicle. The development of the EMS was based on off-the-shelf components and the adoption of a Robot Operating System (ROS), a meta-operating system developed for robotic-oriented applications. Our approach offers soft real-time control and the ability to organize the controller of the EMS as a straightforward and comprehensive message system that provides the necessary inter-process communication at the core of the EMS control procedure. We describe in detail the software-based implementation and validate our approach through experimental results obtained while the prototype was racing in a low-energy consumption competition.

An Overview of Microkernel Based Operating Systems

The creation of Operating Systems (OSs) with Microkernels was in response to the various challenges presented by Operating Systems with Monolithic kernels. Microkernel based Operating Systems provides security and flexibility in the system. This paper reviews seven different microkernel-based Operating Systems: L4, GNU Hurd, Genode, L4re, NOVA, seL4, and Muen Separation Kernel. This analysis provides an understanding of the various trends in the Microkernel Based Operating Systems design. Research papers and official documentation of the individual microkernels served as data sources. The result in this paper shows that there has not been a significant variation in the underlying principle of minimality and how Microkernel Operating Systems approach the implementations of their inter-process communication (IPC), memory management, and scheduling.

A Hardware-in-the-Loop-on-Chip Development System for Teaching and Development of Dynamic Systems

This paper proposes a low-cost on-chip Hardware-in-the-Loop (HIL) platform for teaching and fast prototyping of dynamic systems. A dual-core digital signal controller (DSC)-based solution is proposed for the HIL system. CPU core A, as the simulation engine, is dedicated to circuit and system simulation. The actuation and control logic are implemented in CPU core B, which is working as the control engine. Inter-processor communication is used to interchange variables between the CPUs. The digital-to-analog converter and digital outputs are used to send the duty cycle and system state variables to the oscilloscope for users’ visual feedback. Two typical systems with fast and slow dynamics are modeled and implemented in the simulation engine. Under the excitation generated by the control engine, system dynamics can be observed for studying purposes. Close-loop control for a buck converter is also demonstrated on the developed prototype, where both input voltage and load variations performance are tested. The test results indicate that the digital simulator can well emulate the average small signal model of a power converter in open-loop and close-loop scenario. Meanwhile, the control parameters can be modified for system performance evaluation and education purposes. The proposed low-cost HIL system can be easily applied to the engineering teaching as well as fast prototype development phase of product design.

A First-Order Logic verification framework for communication-parametric and time-aware BPMN collaborations

The BPMN standard notation allows business process designers to model both intra-organizational processes and inter-organizational collaborations. A great effort has been devoted in proposing formal semantics for BPMN, and, fewer, in providing dedicated verification tools. Still, some advanced features of BPMN, namely communication or time-related constructs, are often set aside. This becomes an issue as BPMN gains interest outside of its original scope, e.g., for the IoT where communication and time play an important role. In this paper, we propose a formal semantics for a subset of BPMN. This semantics takes into account not only the usual gateways, but also sub-processes, inter-process communication, and time-related constructs. In contrast to transformational approaches, which give a semantics to BPMN by mapping it to some formal model (e.g., transition systems or Petri nets), our approach is based on a direct formalization in first-order logic that is then realized in a straightforward way into the TLA+ formal language. We build on the TLA+ model-checker, TLC, to provide process designers with a verification framework, fbpmn, that one may use to check BPMN and workflow specific properties. Our tools and our model database are open source and freely available online.

Quality and energy optimized scheduling technique for executing scientific workload in cloud computing environment

<p class="Abstract">Modern BigData data-intensive and scientific workload execution is challenging. The major issues are reliable processing, performance efficiency and energy efficacy perquisite of BigData processing framework. This work assume self-aware MC architectures that autonomously adjust or optimize their performance to accommodate users quality of service (QoS) performance requirement, job execution performance, energy efficiency, and resource accessibility. Extensive workload scheduling has been presented to minimize energy consumption in cloud computing (CC) environment. However, the existing workload scheduling model induces higher amount of interaction cost between inter-processors communications. Further, due to poor resource utilization, routing inefficiency these existing model induces higher energy cost and fails to meet workload QoS prerequisite. For overcoming research challenges, this paper presented quality and energy optimized scheduling (QEOS) technique for executing data-intensive workload by employing dynamic voltage and frequency scaling (DVFS) technique. Experiment outcome shows QEOS model attains good trade-off between system performance and energy consumption in multi-core cloud computing (CC) architectures when compared with existing model.</p>

Reversible and Plausibly Deniable Covert Channels in One-Time Passwords Based on Hash Chains

Covert channels enable stealthy communications over innocent appearing carriers. They are increasingly applied in the network context. However, little work is available that exploits cryptographic primitives in the networking context to establish such covert communications. We present a covert channel between two devices where one device authenticates itself with Lamport’s one-time passwords based on a cryptographic hash function. Our channel enables plausible deniability jointly with reversibility and is applicable in different contexts, such as traditional TCP/IP networks, CPS/IoT communication, blockchain-driven systems and local inter-process communications that apply hash chains. We also present countermeasures to detect the presence of such a covert channel, which are non-trivial because hash values are random-looking binary strings, so that deviations are not likely to be detected. We report on experimental results with MD5 and SHA-3 hash functions for two covert channel variants running in a localhost setup. In particular, we evaluate the channels’ time performance, conduct statistical tests using the NIST suite and run a test for matching hash values between legitimate and covert environments to determine our channels’ stealthiness.

Improving in-memory file system reading performance by fine-grained user-space cache mechanisms

Nowadays, as the memory capacity of servers become larger and larger, distributed in-memory file systems, which enable applications to interact with data at fast speed, have been widely used. However, the existing distributed in-memory file systems still face the problem of low data access performance in small data reading, which seriously reduce their usefulness in many important big data scenarios. In this paper, we analyze the factors that affect the performance of reading in-memory files and propose a two-layer user space cache management mechanism: in the first layer, we cache data packet references to reduce frequent page fault interruptions (packet-level cache); in the second layer, we cache and manage small file data units to avoid redundant inter-process communications (object-level cache). We further design a fine-grained caching model based on the submodular function optimization theory, for efficiently managing the variable-length cache units with partially overlapping fragments on the client side. Experimental results on synthetic and real-world workloads show that compared with the existing cutting-edge systems, the first level cache can double the reading performance on average, and the second level cache can improve random reading performance by more than 4 times. Our caching strategies also outperform the cutting-edge cache algorithms over 20% on hit ratio. Furthermore, the proposed client-side caching framework idea has been adopted by the Alluxio open source community, which shows the practical benefits of this work. • A packet-level cache policy for reducing page faults during sequential data reading. • An in-memory cache layer to cache the variable-length hot file fragments. • Proved that the variable-length cache problem is a submodular optimization problem theoretically. • Comprehensive performance evaluation of the client-side cache strategies.

Key Agreement Over Inter-Process Communication

Today’s computer is often infected by malwares and conventional communication channels such as inter-process communication (IPC) are attractive attack surface for attackers because important information such as user’s personal data and passwords are transmitted between processes over IPC. In addition, there is no other protection other than the access control mechanism provided by the underlying OS, but it is not always sufficient. To improve the situation, this paper proposes a key agreement protocol between processes using a network socket, which is one of the IPC methods. Our protocol provides a means for legitimate processes to cryptographically communicate over the IPC. We use an uncertain channel for secure key agreement over IPC and we found that the IPC channel behaves as the uncertain communication channel due to the process scheduling of the OS. The proposed protocol is based on random number sharing using the messages that the attacker probabilistically fails to obtain and attacker detection who interrupts the protocol. Our protocol provides secure key sharing against an attacker that interrupts the protocol and impersonates legitimate processes. We experiment on the behavior of the uncertain channel on an actual device and confirm that our protocol achieves 128-bit security in a realistic execution time within 8.5 ms. To our best knowledge, our proposal is the first countermeasure for IPC with cryptographic strength under reasonable assumptions.

Towards Optimal Matrix Partitioning for Data Parallel Computing on a Hybrid Heterogeneous Server

Optimal partitioning of a square computational domain over several heterogeneous processors, balancing the load of the processors and minimizing the inter-processor communication cost, is crucial for data parallel dense linear algebra and other applications having similar communication pattern on modern hybrid servers. Although a solution has been found for two processors, the cases of three and more processors are still open. The state of-the-art solution for three processors uses an approximation communication cost function which fails to accurately account for the total amount of data moved between processors, leaving thus the question of its global optimality unanswered. In this work, we formulate and solve a mathematical problem of optimal partitioning a real-valued square over three heterogeneous processors using a new cost function, which accurately accounts for the total amount of data communicated between processors. We also develop an original method for accurate experimental evaluation of the communication time of data movement between memories of the compute devices in the hybrid platform during the execution of data parallel applications. We successfully use this method in the experimental validation of our mathematical results. Finally, we propose a communication energy model predicting the dynamic energy consumption of data movement between processors and experimentally validate its accuracy. This model predicts, and the experiments confirm, that the performance-optimal partition is not necessarily energy optimal.

WestBox: An object-oriented software component for WEST CODAC

Abstract COntrol, Data Acquisition and Communication (CODAC) real-time software codes are key elements for the operation of a fusion device both for the machine protection and for the optimization of the experiments. In 2013, following the WEST (W -for tungsten- Environment Steady-state Tokamak) upgrade of Tore-Supra, the whole legacy acquisition system has been re-factored. The WEST CODAC framework which inherited the older implementations used over 20 years on TORE Supra has been reorganized into primary software components corresponding to basic technical features, such as Database Access, Inter-process Communications, Real-time Device Management, Timing System Network, Finite-State Machine, and Shared Memory Network. In addition, a new modern middleware component named Westbox has been developed, to simplify developments for native diagnostic upgrades and cross-platform data acquisition hardware and software from guest collaborators.