The complexity of large-scale parallel systems necessitates the simultaneous optimization of multiple hardware and software components to meet performance, efficiency, and fault-tolerance goals. A codesign methodology...
详细信息
The complexity of large-scale parallel systems necessitates the simultaneous optimization of multiple hardware and software components to meet performance, efficiency, and fault-tolerance goals. A codesign methodology using modeling can benefit systems on the path to exascale computing.
A classification method for parallel sorting architectures is presented for comparison of existing designs. Some common drawbacks of these parallel sorters are noted: in general, they are limited by their data accessi...
详细信息
A classification method for parallel sorting architectures is presented for comparison of existing designs. Some common drawbacks of these parallel sorters are noted: in general, they are limited by their data accessing mechanism, their hardware utilization is low, and their sorting speed is in some sense independent of the number of sorting elements used in the architecture. In this paper we present a detailed design and analysis of distributor based sorters (parallel balanced tree sorters) which can sort lists whose size is proportional to the memory space available and whose speed is proportional to the number of sorting elements used.
Nomadicity is a new challenge for computing and communication technologies. Modern cellular telephone systems extend the usability of portable personal computers enormously. A nomadic user can be given ubiquitous acce...
详细信息
Nomadicity is a new challenge for computing and communication technologies. Modern cellular telephone systems extend the usability of portable personal computers enormously. A nomadic user can be given ubiquitous access to remote information stores and computing services. However, the behaviour of wireless links creates severe inconveniences within the traditional data communication paradigm. In this paper we give an overview of the problems related to wireless mobility We also present a new software architecture for mastering the problems and discuss a new paradigm for designing mobile distributed applications. The key idea in the architecture is to place a mediator, a distributed intelligent agent, between the mobile node and the wireline network.
The coupling of process-centered environments (PCEs) with distributed object architectures like COM, CORBA and Java opens up an unprecedented range of possibilities in terms of automatic process management. The human ...
详细信息
The coupling of process-centered environments (PCEs) with distributed object architectures like COM, CORBA and Java opens up an unprecedented range of possibilities in terms of automatic process management. The human factor is still the major component of the software precess but its role is changing, as is the kind of support it needs. Focusing on the interactions between different distributed PCEs (heterogeneous federation), we present an Internet-based framework for the definition and enactment of federated processes. A CSP-like process definition language makes distribution problems transparent to the process designer while compiler and execution engines exploit Internet technology in order to map process specifications into the actual IT infrastructure. Different solutions are investigated for both the deployment of the enactment components and their interaction with local PCEs.
New application types such as distributed multimedia applications have to provide a certain quality of service (QoS) to the users. Since they handle time-critical information such as audio and video data, they need ap...
详细信息
New application types such as distributed multimedia applications have to provide a certain quality of service (QoS) to the users. Since they handle time-critical information such as audio and video data, they need appropriate support from the system components a nd especially from the network. New protocols and mechanisms have been developed over recent years to offer integrated services by serving both discrete media data (such as text and graphics) and continuous-media data (i.e, audio and video) in digital networks. Internet and asynchronous transfer mode (ATM) are the main players in this area and both possess QoS architectures which allow them to integrate services of data- and tele- communications formerly performed by separate infrastructures. We believe that both will co-exist for a significant amount of time, potentially complemented by other, perhaps simpler, approaches which are currently under investigation, such as differentiated services. Therefore, an interaction between these two architectures is necessary. In this paper, we discuss interaction approaches for the QoS architectures developed for the Internet and for ATM. We base this description on requirements and scenarios of multimedia applications and on the possible communication patterns considering different topological variants for heterogeneous Internet-ATM networks.
The trend towards ever-larger WWW sites with hundreds of thousands of pages maintained by teams of developers has made apparent the need for tools to manage such large-scale efforts. In this paper we present a mechani...
详细信息
The trend towards ever-larger WWW sites with hundreds of thousands of pages maintained by teams of developers has made apparent the need for tools to manage such large-scale efforts. In this paper we present a mechanism for organising and manipulating groups of nodes and links in WWW sites. These constructs are used to create dynamic views of the data in a given site according to various organisational and presentational criteria. Moreover, proven techniques for the management of hypertext networks (e.g. versioning, variants, etc.) can be readily implemented through the use of this model. Additionally, the model allows the creation of user views through the combination of these constructs via a number of operations. After the description of our model we present a prototype system developed at the University of Geneva that demonstrates how the mechanisms included in our model can be integrated into an existing http server. We then examine how the same mechanisms can be employed in the management of large WWW sites.
A scalable and fault tolerant restricted shared memory architecture is described which uses dual-port memory blocks for interprocessor communication. For an n processor system, n(n - 1)/2 dual-port memory blocks are o...
详细信息
A scalable and fault tolerant restricted shared memory architecture is described which uses dual-port memory blocks for interprocessor communication. For an n processor system, n(n - 1)/2 dual-port memory blocks are organised in the form of a triangular matrix. Multiple processors can communicate concurrently as unique and dedicated dual-port memory blocks connect different pairs of processors.
Current ultra-high-performance computers execute instructions at the rate of roughly 10 PFLOPS (10 quadrillion floating-point operations per second) and dissipate power in the range of 10 MW. The next generation will ...
详细信息
Current ultra-high-performance computers execute instructions at the rate of roughly 10 PFLOPS (10 quadrillion floating-point operations per second) and dissipate power in the range of 10 MW. The next generation will need to execute instructions at EFLOPS rates-100x as fast as today's-but without dissipating any more power. To achieve this challenging goal, the emphasis is on power-efficient execution, and for this we propose VLIW-CMP as a general architectural approach that improves significantly on the power efficiency of existing solutions. Compared to manycore architectures using simple, single-issue cores, VLIW-CMP reduces both power and die area, improves single-thread performance, and maintains aggregate FLOPS per die. To improve further on the power advantages of VLIW, we describe a mechanism that reduces power dissipation of both data forwarding and register-file activity.
This paper presents an original cloud computing architecture for music composition. In this model, music applications are built by making several computer music services work together. Component services are provided ...
详细信息
This paper presents an original cloud computing architecture for music composition. In this model, music applications are built by making several computer music services work together. Component services are provided by a dedicated layer in the cloud architecture called computer music as a Service (CMaaS). The specialized music services can be integrated into different applications at the same time. These music services provided by the CMaaS layer are implemented in the form of platform images based on templates at the Platform as a Service layer. The images are ready to be loaded into the virtualized infrastructure on demand. As examples of implementation over the proposed cloud architecture, two powerful applications for computer music composition are presented: "Diatonic Composer", an interactive composer of scores with high-abstraction music elements, and "Csound Meets the Cloud", an assisted algorithmic composer focused on sound synthesis. The composition model, the involved music services and the web application are described for the above mentioned applications. The proposed architecture, the implemented services and the provided application examples constitute a decided step towards distributed music computation. (C) 2012 Elsevier Ltd. All rights reserved.
Two main areas of activity can be identified in the field of network security: the local area network and the common point between networks. Presents an overview of two projects concerning the above-mentioned areas. W...
详细信息
Two main areas of activity can be identified in the field of network security: the local area network and the common point between networks. Presents an overview of two projects concerning the above-mentioned areas. While the first will concern security architectures based on active firewall components, the second presents a secure intranet solution. The combination of active firewall components with the DMZ concept can translate a security policy into reality and will result in a high level security firewall system. One of the most important security problems in an intranet is to ensure that every user has access only to the information which is relevant for his/her work. In our project this is solved using role-dependent information access for each user.
暂无评论