Giacomo Piscitelli

Performance evaluation of Westwood+ TCP congestion control

Westwood+ TCP is a sender-side only modification of the classic Tahoe/Reno TCP that has been recently proposed to improve fairness and efficiency of TCP. The key idea of Westwood+ TCP is to perform an end-to-end estimate of the bandwidth available for a TCP connection by properly counting and filtering the stream of ACK packets. This estimate is used to adaptively decrease the congestion window and slow-start threshold after a congestion episode. In this way, Westwood+ TCP substitutes the classic multiplicative decrease paradigm with the adaptive decrease paradigm. In this paper we report experimental results that have been obtained running Linux 2.2.20 implementations of Westwood+, Westwood and Reno TCP to ftp data over an emulated WAN and over Internet connections spanning continental and intercontinental distances. In particular, collected measurements show that the bandwidth estimation algorithm employed by Westwood+ nicely tracks the available bandwidth, whereas the TCP Westwood bandwidth estimation algorithm greatly overestimates the available bandwidth because of ACK compression. Live Internet measurements also show that Westwood+ TCP improves the goodput w.r.t. TCP Reno. Finally, computer simulations using ns-2 have been developed to test Westwood, Westwood+ and Reno in controlled scenarios. These simulations show that Westwood+ improves fairness and goodput w.r.t. Reno.

Web applications design and maintenance using symbolic model checking

We present an approach to automatic checking of the correctness of Web applications structure during their lifecycle. The approach adopts the well-established symbolic model checking technique and the associated tool SMV. A formalism allows the designer to describe the model of a Web-based system. Computation Tree Logic (CTL) is adopted as language to define the properties to be verified. The approach has been implemented in a tool to provide automatic support in the design of Web applications. The system embeds the NuSMV model checker to perform verification. Verification is carried out after building the finite state model of a site in the model checker input language. To this purpose the system parses the HTML source code of Web pages, including code for dynamic pages. Properties are expressed using a user friendly interface for Web application developers that automatically translates properties in CTL formulas.

Live Internet measurements using Westwood+ TCP congestion control

Westwood+ TCP is a sender-side only modification of the classic TCP that is based on end-to-end estimation of the bandwidth available to the connection in order to shrink adaptively the TCP congestion window and the slow start threshold after congestion. We report measurements obtained using Linux implementations of Westwood+, Westwood and Reno to FTP data over Internet connections spanning continental and intercontinental distances. In particular, we show that the bandwidth estimation algorithm employed by Westwood+ nicely tracks the available bandwidth, whereas the previous bandwidth estimation algorithm used by TCP Westwood fails to work in the real Internet due to ACK compression. Collected live Internet measurements also show that Westwood+ improves the goodput, from 10% to 46%, with respect to TCP Reno.

AnWeb: a system for automatic support to web application verification

In this paper we propose a formal method for web applications verification. The verification process is carried out by checking that either the system always satisfies a model of the specifications or by producing a counter-example.We represent the system as a Kripke structure and model a web site as a graph. Model checking is reformulated as checking that each initial state satisfies the specifications. We adopt Computation Tree Logic (CTL) as language to define the properties to be verified.The proposed formal method has been deployed in AnWeb, a tool for automatic support in the design of web applications. The tool provides an interface to the SMV model checker. The system parses the HTML source code of web pages, including code for dynamic pages, builds the model in SMV input language and provides the proper CTL specifications to the SMV tool.

Heuristic scheduling of jobs on a multi-product batch processing machine

In this paper we study a scheduling problem on a batch processing machine, i.e. on a processor that can manufacture a set of jobs simultaneously. The scheduling objective is to determine the batches of jobs of different types and their loading sequence to fully utilize the processor and to minimize the tool replacing and setting up times. The schedule is subject to some physical and technological constraints due both to the machine capacity and to the limited number of tools at its disposal. The attention given to this problem is motivated by a recurrent scheduling problem occurring in the production management of a typical shoe factory, which is considered as a case study. The paper gives an efficient heuristic procedure leading to a solution close to the optimal one.

Knowledge based approach to semantic composition of teams in an organization

Finding rapidly suitable experts in an organization to compose a team able to solve specific tasks is a typical problem in large consulting firms. In this paper we present a Description Logics approach to the semantic-based composition of ad-hoc teams based on individuals skill profiles and on task description. The selection process is carried out using a novel Concept Covering algorithm that exploits the recently proposed Concept Abduction inference service in Description Logics. The approach has been deployed as part of a skill management system that takes text files with curricula and project specifications as inputs and extracts from them available individual profiles and task descriptions, according to an ontology modeling skills.

An Agency for Semantic-Based Automatic Discovery of Web-Services

With the evolution of Web service technology, services will not only become increasingly sophisticated, but also move into the area of business-to-consumer and peer-to-peer interactions. Because of todays wide variety of services offered to perform a specific task, there is a need for mediation infrastructures able to support humans or agents in the eventual selection of appropriate services. It is a common opinion that such issues should be solved adopting semantically rich unambiguous descriptions. Hence, ontologies should be used to describe services, to ease their discovery and selection. In order to perform such a selection, a matchmaking procedure, based on semantic descriptions similarity, is needed. Technologies developed for the Semantic Web based on theoretical studies on Artificial Intelligence, particularly on Description Logics, can help in this sense. As the Semantic Web is conceived as an extension of the current one, technologies developed explicitly for they both must be used synergically in order to provide a semantic layer to approaches such as UDDI registries, using OWL formatted descriptions. In this paper we present a framework for discovery of services, stored in an UDDI registry, which exposes a description whose semantic can be modeled using OWL-DL based formalism. In order to perform this task, methodologies to compute semantic differences between two descriptions and non-standard inference services have been investigated and exploited in an implemented system.

Semantic Based Collaborative P2P in Ubiquitous Computing

We propose a collaborative environment for semantic-enabled mobile devices (e.g. PDAs, cell phones, laptops) in peer to peer scenarios. Within the environment, resource discovery is performed exploiting technologies and techniques for knowledge representation developed for the semantic Web, which have been adapted to cope with the highly flexible structure of ad-hoc networks in ubiquitous computing. The approach exploits the standard Bluetooth stack, using the original UUID payload, to carry semantically annotated data. The environment is motivated and presented in a museum case study.

A semantic-based mobile registry for dynamic RFID-based logistics support

In this paper we propose an extended version of the open source jUDDI implementation by the Apache Software Foundation, adapted to pervasive RFID contexts. The registry adopts an OWL-S 1.1 Profile instance annotation of mobile services and resources. Ontology-based metadata are exploited in order to perform a semantic-based service discovery w.r.t. a given request. The proposed framework has been devised for pervasive RFID-based logistics environments. A case study is presented along with experimental results on a prototype implementation.

RFID meets bluetooth in a semantic based u-commerce environment

We present a novel resource discovery framework for u-commerce. Both the original RFID data exchange protocol and the Bluetooth Service Discovery Protocol have been extended in order to enable support for the semantic annotation of products and goods. Given a request, the approach we propose allows an enhanced discovery process exploiting the semantics of resource descriptions available in a u-marketplace. The enhancement is backward compatible with both the original discovery protocols, thus allowing the smooth coexistence of the resource discovery and/or identification approaches. We present and motivate our approach in an innovative u-commerce framework, and show its benefits.