1,720,991 research outputs found
PariPari: Design and Implementation of a Resilient Multi-Purpose Peer-to-Peer Network
Full text linkRecent years have seen a considerable and constant growth in peer-to-peer (P2P) traffic over Internet. Internet Service Providers (ISPs) and software houses have begun to change their attitude towards P2P applications, no longer seen as bandwidth-eager enemies, but rather as interesting profit opportunities.
PariPari is a P2P platform under development at Department of Information Engineering
Engineering of the University of Padova. It differs from traditional P2P applications like eMule, Skype or Azureus in that it provides a multifunctional, extensible platform on which multiple services - from filesharing to VoIP to mail/web/IRC services --- can run simultaneously and cooperatively. PariPari offers a collection of APIs allowing third party developers to write their own applications; but unlike similar P2P development frameworks such as JXTA, PariPari already works ``out of the box'' for the end user offering a large number of applications.
The number and the heterogeneity of services offered by PariPari together with the possibility of extending this platform with future, not currently defined, applications offer a number of challenges: effective use of Java, coordination of multiple services, design of a powerful but easy to use GUI, efficient and robust algorithms for clock synchronization and search etc.
Effective group management was the key to successful development of PariPari.
Over the past three years more than one hundred students have cooperated developing PariPari. To coordinate such a large number of people we have adopted software engineering techniques such as eXtreme Programming and Test Driven Development. However, these paradigms had to be adapted to a group of non-professional, although motivated, developers.
This coordination process was difficult, but extremely rewarding, and taught us a number of lessons about software engineering that might be useful in other software projects involving large numbers of relatively inexperienced, part-time developers with high turnover.Negli ultimi anni il traffico dovuto al peer-to-peer (P2P) è aumentato costantemente. Gli Internet Service Providers (ISPs) e le software house hanno iniziato a considerare le applicazioni P2P non come programmi avidi di banda ma come interessanti opportunità.
PariPari è una piattaforma P2P in sviluppo al Dipartimento di Ingegneria dell'Informazione. È molto diversa da altre ben note applicazioni P2P come eMule, Skype o Azureus. dato che fornisce una piattaforma multifunzionale e estensibile sulla quale diversi servizi --- dal filesharing al VoIP all'email --- possono funzionare simultaneamente.
Inoltre, PariPari fornisce un insieme di API utili agli sviluppatori terzi per scrivere le loro applicazioni, ma diversamente dagli altri framework P2P come JXTA, PariPari offre già da subito un gran numero di applicazioni fruibili dall'utente finale.
PariPari offre, quindi, un gran numero di servizi eterogenei e la possibilità di estendere la piattaforma, in futuro, con applicazioni non ancora definite. Per produrre questi due risultati, la progettazione di PariPari ha dovuto affrontare diverse interessanti sfide tra cui un uso efficiente di Java, la possibilità di coordinare diversi servizi e la studio di nuovi algoritmi per la sincronizzazione e la ricerca.
La chiave del successo dello sviluppo di PariPari è sicuramente la gestione del gruppo.
Negli ultimi tre anni, più di un centinaio di studenti hanno lavorato allo sviluppo di PariPari. Per coordinare tanti contributi abbiamo adottato tecniche tipiche dell'ingegneria del software come l'eXtreme Programming e il Test Driven Development. Questi paradigmi, tuttavia, hanno subito pesanti modifiche per essere adattati al nostro gruppo di sviluppatori dalle peculiari caratteristiche: gli studenti, sebbene motivati, non hanno nè il rendimento nè la preparazione di un professionista.
La gestione è stata complessa ma estremamente appagante ed ha prodotto molti interessanti spunti che possono essere studiati ed applicati ad altri progetti che coinvolgono molti sviluppatori non professionisti con alto turn-over
psort, yet another fast stable sorting software
No full textpsort is the fastest sorting software according to the PennySort benchmark, sorting 181GB of data in 2008 and 224GB in 2009 for $0.01 of computer time. This article details its internals, and the careful fitting of its architecture to the structure of modern PC-class platforms, allowing it to outperform state-of-the-art sorting software such as STXXL sort
psort, yet another fast stable sorting software
No full textpsort was the fastest sorting software in 2008 according to the Pennysort benchmark, sorting 181GB of data for 0.01$ of computer time. This paper details its internals, and the careful fitting of its architecture to the structure of modern PCs-class platforms, allowing it to outperform state-of-the-art sorting software such as GNUsort or STXXL
Obtaining Performance Measures through Microbenchmarking in a Peer-to-Peer Overlay Computer
No full textThe availability of enormous amounts of unused computing power and data storage over the Internet makes the development of a globally distributed computing platform, called Overlay Computer (OC), attractive for the industry, the scientific community, and
also the general user. In order to be effective, an OC
must be equipped with tools for estimating and optimizing the performance of applications. The tools must rely
upon performance metrics of the key factors impacting
performance of a distributed application, such as interconnection latency and bandwidth, and available computing power at the nodes. This paper first reviews the state-of-the-art on performance benchmarking approaches
in globally distributed systems, including performance-
conscious P2P systems. Later, it addresses the problem of developing a suite of microbenchmarking experiments aimed at providing the basic functionalities of a
measurement tool for a P2P-based OC platform. Results show that such a measuring system should take
into account the communication patterns generated by
the applications in order to provide useful performance
insights
Psort, yet another fast stable sorting software
No full textpsort is the fastest sorting software according to the PennySort benchmark, sorting 181GB of data in 2008 and 224GB in 2009 for $0.01 of computer time. This article details its internals, and the careful fitting of its architecture to the structure of modern PC-class platforms, allowing it to outperform state-of-the-art sorting software such as STXXL sort
psort
Full text linkpsort è stato il più veloce software di ordinamento per macchine di classe PC dal 2008 al 2011 (benchmark Pennysort, http://sortbenchmark.org) e un suo adattamento per cluster ha migliorato il record per il benchmark datamation di quasi un ordine di grandezza nel 2011. Il rapporto tecnico ufficiale si trova sul sito sortbenchmark.org (che cataloga i più efficienti software di ordinamento per varie categorie di task/hardware - originariamente mantenuto dal premio Turing Jim Gray) all'URL http://sortbenchmark.org/psort_2011.pdf -- Ulteriori dettagli si possono trovare nelle pubblicazioni:
P. Bertasi, M. Bressan, E. Peserico. psort, yet another fast stable sorting software, ACM Journal of Experimental Algorithmics, vol. 16, 2011 --
P. Bertasi, M. Bonazza, M. Bressan, E. Peserico. Datamation: a quarter of a century and four orders of magnitude later. Proc. of IEEE CLUSTER 201
A Novel Resource-Driven Job Allocation Scheme for Desktop Grid Environments
No full textIn this paper we propose a novel framework for the dynamic allocation of jobs in grid-like environments, in which such jobs are dispatched to the machines of the grid by a centralized scheduler. We apply a new, full resource-driven approach to the scheduling task: jobs are allocated and (possibly) relocated on the basis of the matching between their resource requirements and the characteristics of the machines in the grid. We provide experimental evidence that our approach effectively exploits the computational resources at hand, successfully keeping the completion time of the jobs low, even without having knowledge of the actual running times of the jobs
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
- …
