The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Large-scale graphs can be analyzed by lightweight systems on a single server, e.g., GraphChi, X-Stream, and GridGraph. Studies indicate that graph algorithms have different performance impacted by partitioning schemes, scheduling strategies and execution models. Existing systems of single-server graph processing often suffer from poor I/O locality, inefficient selective scheduling or expensive synchronization...
MeteoSwiss, the Swiss national weather forecast institute, has selected densely populated accelerator servers as their primary system to compute weather forecast simulation. Servers with multiple accelerator devices that are primarily connected by a PCI-Express (PCIe) network achieve a significantly higher energy efficiency. Memory transfers between accelerators in such a system are subjected to PCIe...
Accurate simulation of unsteady turbulent flow is critical for improved design of greener aircraft that are quieter and more fuel-efficient. We demonstrate application of PyFR, a Python based computational fluid dynamics solver, to petascale simulation of such flow problems. Rationale behind algorithmic choices, which offer increased levels of accuracy and enable sustained computation at up to 58%...
Indirect and direct learning architectures are the two main parameter identification approaches for digital predistortion systems. While the indirect scheme is less complex, its inherent shortcomings are avoided by the direct learning approach. Trying to answer the question whether this advantage of the direct approach can be exploited in terms of measurable linearization-performance improvement in...
Reading and writing data efficiently from storage systems is critical for high performance data-centric applications. These I/O systems are being increasingly characterized by complex topologies and deeper memory hierarchies. Effective parallel I/O solutions are needed to scale applications on current and future supercomputers. Data aggregation is an efficient approach consisting of electing some...
Packet-pair technique is a widely used method for characterizing end-to-end network paths. A new analytical model is presented for the packet-pair based signature that accurately describes the behavior of packet-pairs in multihop network paths with multiple tight links. The relationship between the input and output gaps of packet-pairs and the corresponding distribution of end-to-end packet-pair dispersions...
Stencil computation is an important class of algorithms used in a large variety of scientific-simulation applications. The performance of stencil calculations is often bounded by memory bandwidth. High-bandwidth memory (HBM) on devices such as those in the Intel® Xeon Phi™ ™200 processor family (code-named Knights Landing) can thus provide additional performance. In a traditional sequential time-step...
This paper considers the problems of efficiency estimation of using different types of modulation of hydroacoustic signals of AUV communication, navigation and control systems. The types of signal were investigated: simple signal, linear frequency modulation signal and direct sequence spread spectrum. The signals were compared by the noise immunity, time resolution and detection probability with usage...
Energy consumption represents a large percentage of the operational expenses in data centers. Most of the existing solutions for energy-aware scheduling are focusing on job distribution and consolidation between computing servers, while network characteristics are not considered. In this paper, we propose a model of power and network-aware scheduling that can be tuned to achieve energy-savings, through...
The NATO Object Level Protection (OLP) is an approach to Allied data protection that aims to protect individual information objects separately. OLP has an evolution stage which involves the direct use of cryptography in this context, called Cryptographic Access Control. Using next-generation public-key encryption techniques called attribute-based encryption (ABE), RBAC policies may be directly encoded...
Geographically Weighted Regression (GWR) is a local version of spatial regression that captures spatial dependency in regression analysis. GWR has many application in practice as a visualization and prediction tool for spatial exploration (e.g in climate, economy, medical). However, this locally regression model is slow in process upon the volume of calculations and the spatial getting bigger. Improving...
We propose a novel inter-layer path control mechanism in a PCE-VNTM cooperative multi-layer networks (MLN). For this, we define a novel FA-LSP aware PCE which only reports and manages a forwarding adjacency-label switched path (FA-LSP) state. In our scheme, the FA-LSP can be distinguished from the pure higher-layer TE link more and accurately controlled with a network policy. We worked out the methodology...
Hybrid clouds, consisting of multiple individual smaller clouds with heterogeneous capabilities, are becoming more and more popular through concepts such as inter-clouds, fog-and edge-computing. They provide fast computations without introducing large network latency. However, such cloud environments often contain unreliable nodes and links that are failure prone. Therefore, the deployment of applications...
The emerging research of SDN and NFV have been promising to provide flexibility in network provisioning based on service requirement. However, current network provision methods of SDN and NFV do not have a formalized model of describing dynamic requirement of the services. The usage of audio/video services may be the example of the services with dynamic requirement. The orchestrator of SDN and NFV...
To satisfy the QoS requirements and improve the reliability of the network, we propose a load-balancing-based fault-tolerant mapping method (LFMM) in smart grid virtual networks. The process of LFMM is divided into two stages, one is node mapping stage, and the other is link mapping stage. During node mapping stage, we present a load-balancing-based virtual node mapping (LVNM) algorithm. We choose...
In dynamic bandwidth management based on traffic prediction, the traffic flows can be modeled as time-series data. State-of-the-art technique used in modeling this traffic flows is by using a linear model. In contrast, Recurrent Neural Network (RNN) has been the state-of-the-art technique in speech recognition, which data is also time-series. Therefore, we conjecture that the use of RNN can improve...
3D integration provides opportunities to design high-bandwidth and low-power CMOS image sensors (CIS) [1–4]. The 3D stacking of pixel tier, peripheral tier, memory tier(s), and compute tier(s) enables high degree of parallel processing. Also, each tier can be designed in different technology nodes (heterogeneous integration) to further improve power-efficiency. This paper presents a case study of...
An ever increasing utility and use of virtualization in various emerging scenarios, e.g.: Cloud Computing, Software Defined Networks, Data Streaming Processing, asks the Infrastructure Providers (InPs) to optimize the allocation of the virtual network requests (VNRs) into a substrate network. In this paper we present a two-stage virtual network embedding (VNE) algorithm, which map first virtual nodes...
In the last decade, the scope of software optimizations expanded to encompass energy consumption on top of the classical runtime minimization objective. In that context, several optimizations have been developed to improve the software energy efficiency. However, these optimizations commonly rely on long profiling steps and are often implemented as unstable runtime systems, which limits their applicability...
Today, accelerator cards like GPUs are an important constituent of HPC clusters. For certain GPU-intense applications, the trend is shifting toward multi-GPU systems with four or more GPUs per compute node. This can increase the performance per dollar and the performance per watt. The Linpack benchmark is the standard tool for measuring the compute performance of supercomputers. Its standard implementation,...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.