1、OpenMP-based parallelization on an MPCore multiprocessor platform A performance and power analysis Original Research ArticleJournal of Systems Architecture, Volume 54, Issue 11, November 2008, Pages 1019-1029H. Blume, J. von Livonius, L. Rotenberg, T.G. Noll, H. Bothe, J. BrakensiekClose preview | P
2、urchase PDF (739 K) | Related articles | Related reference work articles Abstract | Figures/Tables | References AbstractIn this contribution, the potential of parallelized software that implements algorithms of digital signal processing on a multicore processor platform is analyzed. For this purpose
3、 various digital signal processing tasks have been implemented on a prototyping platform i.e. an ARM MPCore featuring four ARM11 processor cores. In order to analyze the effect of parallelization on the resulting performance-power ratio, influencing parameters like e.g. the number of issued program
4、threads have been studied. For parallelization issues the OpenMP programming model has been used which can be efficiently applied on C-level. In order to elaborate power efficient code also a functional and instruction level power model of the MPCore has been derived which features a high estimation
5、 accuracy. Using this power model and exploiting the capabilities of OpenMP a variety of exemplary tasks could be efficiently parallelized. The general efficiency potential of parallelization for multiprocessor architectures can be assembled.Article Outline1. Introduction2. MPCore architecture3. Par
6、allelization of programs using OpenMP4. Exemplary parallelization of a motion estimation algorithm5. Hybrid FLPA/ILPA modeling of the MPCore6. Evaluation of the MPCore architecture executing parallelized tasks7. ConclusionReferencesVitaeIssues in the development of a practical NoC: the Proteo concep
7、t Original Research ArticleIntegration, the VLSI Journal, Volume 38, Issue 1, October 2004, Pages 95-105David Sigenza-Tortosa, Tapani Ahonen, Jari NurmiClose preview | Purchase PDF (284 K) | Related articles | Related reference work articles Abstract | Figures/Tables | References AbstractNetwork-on-
8、Chip will be one of the cornerstones of future electronics. At Tampere University of Technology we have been working on the development of our own proposal for a flexible on-chip communication network, called Proteo. Proteo introduces the concept of an open library of communication components that c
9、an be selected and configured to build highly-customized networks-on-chip. The designer of a new System-on-Chip platform starts with a description of the hardware components of the system and an abstract model of the problem application, and with the help of the Proteo software tools, obtains a synt
10、hesizable instance of a packet-switching network that, ideally, meets his requirements. The constraints placed on the type of designs that may use Proteo are minimal and an important part of the process should be automated. In this article we introduce the philosophy behind the project in relation t
11、o fundamental deep submicron technology problems, and some of our initial results.Article Outline1. Introduction2. NoC design 2.1. Comparing NoC and LAN2.2. Proteo NoC3. Proteo hardware elements 3.1. Nodes and hubs4. Proteo architecture 4.1. Topology4.2. Packets4.3. Protocols5. Automation6. Conclusi
12、onReferencesFutures of global interdependence (FUGI) global modeling system: Integrated global model for sustainable development Original Research ArticleJournal of Policy Modeling, Volume 27, Issue 1, February 2005, Pages 101-135Akira OnishiClose preview | Purchase PDF (860 K) | Related articles |
13、Related reference work articles Abstract | Figures/Tables | References AbstractThe FUGI (futures of global interdependence) global modeling system has been developed as a scientific policy simulation tool of providing global information to the human society and finding out possibilities of policy co
14、ordination among countries in order to achieve sustainable development of the global economy under the constraints of rapidly changing global environment. The FUGI global model M200 classifies the world into 200 countries/regions where each national/regional model is globally interdependent through
15、international trade, export/import prices, financial flows, ODA, private foreign direct investment, exchange rates, stock market prices and policy information, etc. The latest software of FUGI global modeling system (FGMS200) for the Windows 2000/xp professional is also available.Article Outline1. I
16、ntroduction2. Outline of FUGI global modeling system 2.1. Regional classification2.2. Scientific design concept of the global modeling system2.3. Model structure 2.3.1. Population system2.3.2. Economic development system 2.3.2.1. Labor and production at constant prices2.3.2.2. Expenditures on GDP at
17、 constant prices2.3.2.3. Income distributionprofits and wages2.3.2.4. Price2.3.2.5. Expenditures on GDP at current prices2.3.2.6. Money, interest rates and financial assets2.3.2.7. Government finance2.3.2.8. International balance of payments2.3.2.9. International finance2.3.2.10. Foreign exchange ra
18、tes2.4. Model software3. Some examples of estimated parameters of the model4. The projections of the world economy in the global interdependence up to 20105. Alternative scenario simulations: policy exercises 5.1. Assessment of geographical risks: impacts of the war in Iraq on the global economy 5.1
19、.1. Scenario A-15.2. The US strategy against geographical risks: tax cutbacks 5.2.1. Scenario A-25.3. The US dollar depreciation against EURO and Yen5.4. Strategy against the Japan syndromes 5.4.1. Scenario A: self-reliance of private sectorthis scenario corresponds to the baseline scenario5.4.2. Sc
20、enario B: tax cut policy exercises6. ConclusionAppendix A. AppendixReferencesVitaeModeling and analysis of PC-based software routers Original Research ArticleComputer Communications, Volume 33, Issue 12, 15 July 2010, Pages 1462-1470K. SalahClose preview | Purchase PDF (443 K) | Related articles | R
21、elated reference work articles Abstract | Figures/Tables | References AbstractPC-based software routers are major and common network elements of todays Internet infrastructure. In this paper, we model and analyze the performance of PC-based software routers. Our model adopts realistic assumptions of
22、 using finite queues and multi-phase services. Specifically, we model a router with two interfaces as two finite queueing systems in tandem with the first queueing system being a two-stage service and the second queueing system being a single-stage service. The two-stage service consists of a first
23、stage in which data-link layer functionalities are performed, and a second stage in which forwarding/routing functionalities are performed. In this paper, we verify and validate our analytical model by using simulation and real experimental measurements obtained from a PC router with two interfaces.
24、 In addition, we discuss how our analytical models can be used to analyze a router with multiple interfaces. From our models, we derive equations for key features and performance measures of engineering and design significance. These features and measures include throughput, packet loss, packet dela
25、y, average number of packets in the queues and in the overall system, and CPU utilization and availability.Article Outline1. Introduction2. IP forwarding/routing3. Finite queueing model 3.1. Modeling two-stage finite queueing system3.2. Overall IP forwarding model of two queues in tandem4. Multiple interface router5. Verification and validation6. Results and discussion7. ConclusionAcknowledgementsReferences
