Abstract Despite their dominance and wide acceptance, relational databases are unable to meet all waits and needs of multi-tenant applications. By way of example, relational databases do not provide any “tenant” concept, therefore a typical relational database has to be tweaked to support multi-tenancy. This paper addresses this shortage by presenting a new schema-sharing technique for multi-tenant applications. Our approach is... Despite their dominance and wide acceptance, relational databases are unable to meet all waits and needs of multi-tenant applications. By way of example, relational databases do not provide any “tenant” concept, therefore a typical relational database has to be tweaked to support multi-tenancy. This paper addresses this shortage by presenting a new schema-sharing technique for multi-tenant applications. Our approach is built on top of the Extension table method and makes use of the native XML data support to store the additional data supplied by each tenant. It also demonstrates how we integrated our proposed technique in to a real-world prototype Software as a Service application that was designed to enhance publics access to up-to-date records of drinking water monitoring results. It also discusses the benefits and drawbacks of our new hybrid schema-sharing technique. Read More Read Less

Abstract Nowadays, integrity, availability and reliability from information systems have been threatened by intrusions and Unwanted Internet Traffic (UIT), Intrusion Detection Systems (IDS) are largely employed to cope with UIT, but IDS lack in security as they are mainly based on postmortem approaches: detection and/or blocking happen only after UIT has inflicted serious damage. Intending to improve intrusion detection, in our... Nowadays, integrity, availability and reliability from information systems have been threatened by intrusions and Unwanted Internet Traffic (UIT), Intrusion Detection Systems (IDS) are largely employed to cope with UIT, but IDS lack in security as they are mainly based on postmortem approaches: detection and/or blocking happen only after UIT has inflicted serious damage. Intending to improve intrusion detection, in our earlier work we proposed an approach to cope with UIT in a proactive manner, using forecasting techniques combined with Return on Security Investment (ROSI). In this paper we examine the applicability of a cooperative architecture regarding forecasts of UIT on a more complex set-up, with hosts associated with sites geographically divided. The aim of this paper is to show a cooperative architecture of IDS with prediction approaches, covering the gaps of the current forecasting techniques concerning UIT: sensors employment, the use of just one prediction technique and forecasts’ sharing. A proof of concept of such architecture is presented, which allows concluding about the improvement in forecasts for IDS to deal with UI. Read More Read Less

Abstract In this paper the authors compare Canadian and the United Kingdom practices in the management of electronic information on the Internet. The focus is primarily on policies and practices of electronic data storage and transfer, protection of privacy, and electronic business data transfer. The authors examined the impact that national personal data protection legislation and regulations have on electronic data... In this paper the authors compare Canadian and the United Kingdom practices in the management of electronic information on the Internet. The focus is primarily on policies and practices of electronic data storage and transfer, protection of privacy, and electronic business data transfer. The authors examined the impact that national personal data protection legislation and regulations have on electronic data transfers between and within Canada and the UK by organizing consultations with security and privacy experts in both countries, as well as with Canadian and international legislators. The findings are situated within the Access Rainbow framework suitable for obtaining a global view on the roles of different stakeholders in the adoption of e-commerce. This study represents the first stage of a larger project that will encompass several geographic regions1. Read More Read Less

Abstract It has been demonstrated through simulation and experimental studies that TCP protocol suffers from a significant drop of performance when communicating over wireless networks. In general, packets can be lost for three main reasons: 1) Congestion 2) Link failure due to node mobility, and 3) Losses due to transmission errors. However, TCP doesn’t distinguish between these three types of packet... It has been demonstrated through simulation and experimental studies that TCP protocol suffers from a significant drop of performance when communicating over wireless networks. In general, packets can be lost for three main reasons: 1) Congestion 2) Link failure due to node mobility, and 3) Losses due to transmission errors. However, TCP doesn’t distinguish between these three types of packet losses. It considers any packet loss as a sign of network congestion; and therefore triggers its exponential back off as a mean to control congestion. In this paper, we study the impact of providing TCP with explicit information extracted from the physical, link and network layers to make it possible to identify the cause of the packet loss. There are three information of interest to this study: maximum probability to drop a packet, signal strength of the wireless node, and explicit handover notification when the mobile node transits to a new access point. We show in the experimental section that by providing TCP sender with only the information regarding the queues states, the performance can be improved by 500% over a lossy wireless link (with 60% error rate). Read More Read Less