Filters
total: 100
filtered: 63
Search results for: distributed-computing
-
Centralized and Distributed Structures of Intelligent Systems for Aided Design of Ship Automation
PublicationA design process and accepted solutions made during this process, often base on non-formalized knowledge, obtained from designer (expert) intuition and practice. There are no formalized rules assuring the correctness of design solutions. The analysis of design process of ship automation, including ship power system, shows that this process can be supported by application of the artificial intelligence elements. The article presents...
-
Towards Scalable Simulation of Federated Learning
PublicationFederated learning (FL) allows to train models on decentralized data while maintaining data privacy, which unlocks the availability of large and diverse datasets for many practical applications. The ongoing development of aggregation algorithms, distribution architectures and software implementations aims for enabling federated setups employing thousands of distributed devices, selected from millions. Since the availability of...
-
A Solution to Image Processing with Parallel MPI I/O and Distributed NVRAM Cache
PublicationThe paper presents a new approach to parallel image processing using byte addressable, non-volatile memory (NVRAM). We show that our custom built MPI I/O implementation of selected functions that use a distributed cache that incorporates NVRAMs located in cluster nodes can be used for efficient processing of large images. We demonstrate performance benefits of such a solution compared to a traditional implementation without NVRAM...
-
Qualitative evaluation of distributed clinical systems supporting research teams working on large-scale data
PublicationInthispaper,fivecontemporaryscalablesystemstosupportmedicalresearchteams are presented. Their functionalities extend from heterogeneous unstructured data acquisition through large-scale data storing, to on-the-fly analyzing by using robust methods. Such kinds of systems can be useful in the development of new medical procedures and recommendation rules for decision support systems. A short description of each of them is provided....
-
Investigation into MPI All-Reduce Performance in a Distributed Cluster with Consideration of Imbalanced Process Arrival Patterns
PublicationThe paper presents an evaluation of all-reduce collective MPI algorithms for an environment based on a geographically-distributed compute cluster. The testbed was split into two sites: CI TASK in Gdansk University of Technology and ICM in University of Warsaw, located about 300 km from each other, both connected by a fast optical fiber Ethernet-based 100 Gbps network (900 km part of the PIONIER backbone). Each site hosted a set...
-
Robustness Analysis of a Distributed MPC Control System of a Turbo-Generator Set of a Nuclear Plant – Disturbance Issues
PublicationTypically, there are two main control loops with PI controllers operating at each turbo-generator set. In this paper, a distributed model predictive controller with local quadratic model predictive controllers for the turbine generator is proposed instead of a set of classical PI controllers. The local quadratic predictive controllers utilize step-response models for the controlled system components. The parameters of these models...
-
KernelHive: a new workflow-based framework for multilevel high performance computing using clusters and workstations with CPUs and GPUs
PublicationThe paper presents a new open-source framework called KernelHive for multilevel parallelization of computations among various clusters, cluster nodes, and finally, among both CPUs and GPUs for a particular application. An application is modeled as an acyclic directed graph with a possibility to run nodes in parallel and automatic expansion of nodes (called node unrolling) depending on the number of computation units available....
-
Topology recognition and leader election in colored networks
PublicationTopology recognition and leader election are fundamental tasks in distributed computing in networks. The first of them requires each node to find a labeled isomorphic copy of the network, while the result of the second one consists in a single node adopting the label 1 (leader), with all other nodes adopting the label 0 and learning a path to the leader. We consider both these problems in networks whose nodes are equipped with...
-
A Point Set Connection Problem for Autonomous Mobile Robots in a Grid
PublicationConsider an orthogonal grid of streets and avenues in a Manhattan-like city populated by stationary sensor modules at some intersections and mobile robots that can serve as relays of information that the modules exchange, where both module-module and module-robot communication is limited to a straight line of sight within the grid. The robots are oblivious and move asynchronously. We present a distributed algorithm that, given...
-
Deep learning in the fog
PublicationIn the era of a ubiquitous Internet of Things and fast artificial intelligence advance, especially thanks to deep learning networks and hardware acceleration, we face rapid growth of highly decentralized and intelligent solutions that offer functionality of data processing closer to the end user. Internet of Things usually produces a huge amount of data that to be effectively analyzed, especially with neural networks, demands high...
-
Integration of heterogeneous web services in exceptional situations
PublicationWeb services are intended to enable interoperability between heterogeneous distributed systems. Although the technology has been widely adopted and accepted, there are still differences between runtime platforms in exception structure and handling. This results in difficulties in effective handling of exceptions during Web services invocation. The paper presents a solution that enables coordinated exception handling between different...
-
Modelling and simulation of GPU processing in the MERPSYS environment
PublicationIn this work, we evaluate an analytical GPU performance model based on Little's law, that expresses the kernel execution time in terms of latency bound, throughput bound, and achieved occupancy. We then combine it with the results of several research papers, introduce equations for data transfer time estimation, and finally incorporate it into the MERPSYS framework, which is a general-purpose simulator for parallel and distributed...
-
Process arrival pattern aware algorithms for acceleration of scatter and gather operations
PublicationImbalanced process arrival patterns (PAPs) are ubiquitous in many parallel and distributed systems, especially in HPC ones. The collective operations, e.g. in MPI, are designed for equal process arrival times (PATs), and are not optimized for deviations in their appearance. We propose eight new PAP-aware algorithms for the scatter and gather operations. They are binomial or linear tree adaptations introducing additional process...