In this paper, virtual machine performance is determined while stationary and during a live migration process. Benchmarks are used to determine network, CPU, and disk I/O performance of the virtual machines while stationary and migrating. To further examine virtual machine performance, three types of persistent storage techniques are used: Solid-State Drive (SSD), Hard Disk Drive (HDD), and Diskless (Network Boot). Results are given that show the degraded performance of a diskless cloud cluster, due to the network latency of fetch requests located on a networked PXE server. The diskless cloud cluster is compared to the performance of a traditional diskfull cluster consisting of SSDs and HDDs.

Live virtual machine migration allows resources from one physical server to be moved to another with little or no interruption in the processes of the guest operating system. The process involved in performing a live migration includes copying the guest virtual machine memory state and cpu register state from a hyper visor on one server to another. In this paper, virtual machine migration is evaluated in terms of performance of the virtual machine while being migrated, performance of the cloud architecture while migrating the virtual machine, and the power costs of performing a live virtual machine migration. The experiments on virtual machine live migration will give insight on how the cloud architecture responds and handles live migrations, and how virtual machines perform while they are in a migration phase.

Database as a Service is a very attractive product from cloud service providers for small start-up companies. However, consistency maintenance among replica servers is a very big issue in cloud databases. Too much interdependency of replica servers in the existing consistency model reduces performance and throughput, and can increase the transaction failure rate of a cloud database. In this paper we propose a tree-based consistency approach that reduces interdependency among replica servers by introducing partially consistent and fully consistent states of cloud databases. The tree is formed such a way that the maximum reliable path is ensured from the primary server to all replica servers. As a result, the probability of a transaction failure is greatly reduced, which helps to increase performance and throughput even in an unreliable network.

While data grids can provide the ability to solve large-scale applications which require the processing of large amounts of data, they have been recognized as extremely energy inefficient. Computing elements can be located far away from the data storage elements. A common solution to improve availability and file access time in such environments is to replicate the data, resulting in the creation of copies of data files at many different sites. The energy efficiency of the data centers storing this data is one of the biggest issues in data intensive computing. Since power is needed to transmit, store and cool the data, we propose to minimize the amount of data transmitted and stored by utilizing smart replication strategies that are data aware. In this paper we present a new data replication approach, called the sliding window replica strategy (SWIN), that is not only data aware, but is also energy efficient. We measure the performance of SWIN and existing replica strategies on our Sage green cluster to study the power consumption of the strategies. Results from this study have implications beyond our cluster to the management of data in clouds.

Abstract WiFi is becoming increasingly prevalent nowadays, whether as a simple range extender for a home wired Ethernet interface or a wireless deployment throughout an enterprise. Wireless local area networks (WLANs) provide us with mobility, convenience, and low cost. At the same time, WiFi is unsafe and is more vulnerable than traditional Ethernet, so that anyone familiar with wireless networks can initiate an attack. One strategy to identify potentially malicious unauthorized users is through packet capturing. Although there are several software products available for packet capture, there is currently no paper in the literature that describes how to build a software tool to capture WiFi frames and its associated functions. In this paper, we present how we build our frame capture tool along with a set of implementation techniques for automatically capturing all the frames and analyzing an attack on a WiFi. In our research, we focus on the WiFi medium access control (MAC) layer for wireless network analysis. We also discuss what we learned and the limitations which we discovered when implementing the tool. Copyright © 2009 John Wiley & Sons, Ltd.

Abstract The open nature of wireless medium has left wireless communications vulnerable to various privacy attacks. Much research work has been proposed to protect the identity anonymity of communicating parties, the anonymity of network routes and the location privacy of the message source and destination. However, with the advent of new radio identification and localization techniques, more advanced privacy attacks are possible. We describe a new privacy attack in which the adversary tries to infer the itineraries of the nodes in the network. To protect itinerary privacy, we design an algorithm, called$\Delta$ ‐camouflaging mobility algorithm, which changes the original motion segments of a node into$\Delta$ ‐shaped camouflaging paths. Itinerary privacy, correspondent privacy, and route privacy are closely related so we propose a comprehensive anonymous routing scheme, calledMARS, to protect these privacy goals. MARS exploits camouflaging mobility to protect itinerary privacy and usesmotion pseudonymsgenerated from the camouflaging mobility to protect correspondent privacy and route privacy without using cryptography. Our analysis results show that$\Delta$ ‐camouflaging mobility algorithm is cost‐effective, which can significantly reduce the itinerary exposure probability at a small cost of extra travel distance. Moreover, our network simulation results illustrated that MARS anonymous routing scheme and$\Delta$ ‐camouflaging mobility algorithm did not reduce network layer performance. Copyright © 2009 John Wiley & Sons, Ltd.

We present a collaborated camouflaging mobility algorithm to protect mobile hosts' mobility privacy. Our algorithm alters a mobile host's straight moving paths into Delta-shaped camouflaging paths, which cause its wireless transmissions to be distributed over a large area, and hence, reduce the probability of generating traceable wireless transmissions. Furthermore, on the premise of not affecting the mobile hosts' itineraries, our collaborated mobility algorithm clusters the motion paths of mobile hosts to make it difficult for an adversary to identify the mobile hosts. We model the objective of maximizing mobility privacy and minimizing the overhead (e.g. the extra travel distance) as a nonlinear constrained optimization problem and use the solution to the optimization problem as the collaborated mobility. Lastly, we analyze the calculation time of the collaborated camouflaging mobility and conduct simulations to evaluate the privacy improvements brought by the collaborated camouflaging mobility algorithm compared with non-covered mobility.

In this paper, we discuss how to prevent users' passwords from being stolen by adversaries. We propose a virtual password concept involving a small amount of human computing to secure users' passwords in on-line environments. We adopt user-determined randomized linear generation functions to secure users' passwords based on the fact that a server has more information than any adversary does. We analyze how the proposed scheme defends against phishing, key logger, and shoulder-surfing attacks.

In this paper, we discuss how to prevent users' passwords from being stolen by adversaries. We propose differentiated security mechanisms in which a user has the freedom to choose a virtual password scheme ranging from weak security to strong security. Among the schemes, we have a default method (i.e., traditional password scheme), system recommended function, user-specified function, user-specified program, etc. We further propose a codebook approach to serve as system recommended functions and provide a security analysis. For user-specified functions, we adopt secret little functions, in which security is enhanced by hiding secret functions/algorithms.

Adversaries can easily launch an attack, by taking advantage of the free wireless networking available to the public at many conferences, airports or cafes. Meanwhile in these environments, it is very difficult to identify the adversaries. In this paper, we present a scheme that is designed to defend against potential attacks at a very early stage, in order to minimize the damage to the victims and the entire network. Instead of taking actions only after the attacks arrive at the host, our scheme makes the access point the one who monitors, identifies and prevents potential attacks. In this active protection way, the system can respond more quickly and can also prevent much of the malicious network traffic.

The privacy of wireless communications is becoming an important issue due to the open nature of wireless medium. Much research work has been proposed to address the anonymity of communicating parties, the location privacy of the message source and destination, and the privacy of the network routing paths. However, with the advent of new radio identification and localization techniques, more advanced privacy attacks are possible. We describe a new privacy attack in which the adversary tries to infer the itineraries of the nodes in the network. To protect itinerary privacy, we design an algorithm, called the Delta-mobility camouflaging algorithm, which can be applied upon any mobility model by changing the original motion segments into Delta-shaped camouflaging paths. Our analysis results show that the Delta-mobility camouflaging algorithm is cost-effective, which in most cases decreases the itinerary exposure probability more than 80% at a cost of less than 3% extra travel distance.

In this paper, we carry out a comprehensive study to compare invalidation report (IR) to three other cache access algorithms, including poll-each-read (PER), call-back (CB), and lease schemes in future multiple cell wireless Internet. The purpose of this study is to study scalability of IR schemes. To the best of our knowledge, this is the first such study. We focus on the scalability issue of these four fundamental strong-consistent schemes in terms of network transmission costs regarding network size, database size, subscription ratio, and network traffic through extensive computer simulations. Our results show that: 1) the IR schemes do not perform well in multiple-cell wireless Internet with a large update rate, database size, and IR window size, and with a small subscription density; 2) the IR schemes do scale up well with the IR period and the access rate; however, good performance can be obtained when the IR period is large, which implies a large access latency; 3) the PER CB and the lease schemes perform well with a large update rate, small subscription ratio, and large database size.

Location information of users can now be collected from most wireless communication using advanced wireless location tracking techniques. Providing location information can be advantageous in some situations. However, there are instances, where it may be critical to protect the location of the individual. Several protection strategies, such as periodically updating interface identifiers, have been proposed so that an adversary cannot track mobiles in long-term movements. In this paper, we introduce a new strategy, DMAS (Dynamical Mac Assignment with Shuffle), in which the mobile client dynamically exchanges its assigned Mac addresses with others.We present a security analysis to show this scheme can greatly secure a client's location privacy.

Scheduling strategies for on-demand data in broadcast systems typically consider how to minimize the wait time of the requests. When users' requests for data in a broadcast system have real-time constraints, scheduling strategies for such requests typically only consider how to minimize the number of deadlines missed. There are many applications with both real-time and non-real-time requests that would benefit from a broadcast scheduling strategy that considers both the timing constraints and the wait times of requests. We refer to such a broadcast environment as a mixed-type request broadcast environment. In this paper, we present an on-demand broadcast cost model for mixed-type broadcast environments that considers both the response time and number of deadlines missed. We propose a scheduling strategy for mixed-type broadcast systems, called the maximum paid cost first (MPCF) that is based on this cost model. The simulation results show that our MPCF strategy always achieves the best result for varying request arrival rates, ratio of non-real-time requests and real-time requests, and a weighted missed deadline value, when compared to existing broadcast strategies.

In a data intensive grid system, many data replica schemes and models have been proposed to improve the system response time or data consistency, but little attention has been paid to the system reliability. In this paper, we investigate how these data replica schemes impact the system reliability. We use several metrics of system reliability we previously proposed (system bytes missing rate and system file missing rate), and we model the system availability problem assuming limited replica storage and different sized files. In order to achieve higher system data availability, in this paper we propose two new replica optimizers, MinDmr-beta and MinDmr-gamma, to minimize the data miss rate (MinDmr). A comparison of our replica optimizers using a simulation on the OptorSim demonstrates that by utilizing our strategy, we can increase the data availability for files with different sizes.

In this paper, we propose a novel dynamic Mac address assignment and exchange strategy to reduce location disclosure risks. The use of dynamic Mac addresses in a local area network (LAN) is feasible since the address uniqueness needs only to be handled within the LAN. Though each node has a universal Mac address assigned by the manufacturer, a local unique Mac address is sufficient for the node to participate in communications. Our proposed strategy includes three major schemes. The first is to assign a Mac address to the client using the same idea as the IP address assignment. Next, a Mac addresses shuffle is conducted among the clients in the wide local area network (WLAN) without exposing any exchange relations or frequently disrupting current connections. Lastly, dummy messages are sent from the wireless client to cover others during idle. In current wireless networking, a client must communicate with an access point (AP) and authorization server in order to obtain authorization to access a local wireless network. In our scheme, we do not use a node's real physical Mac address when associating with an AP. Instead, there is a public special Mac address broadcast in the beacon messages which will be chosen as the source Mac address in the association request frame. As a result, the user will never expose its identifier when associated with the AP. The client also keeps a variant silent period as a defense against a correlation attack.

The data availability in a data grid system is complicated by node failure, data catalog error and an unreliable network. To improve the job response time and data availability, data is typically replicated in large scale data-massive applications. However, the dynamic behavior of a Grid user makes it difficult to determine where and how to make data replications to meet the system availability goal. Some strategies for data replication have previously been proposed, but they assumed unlimited storage for replicas. In this paper, we present two new metrics to measure the system data availability. We then model the system availability problem assuming limited replica storage and transfer this to a classic optimal problem. We present four strategies for limited replica storage that maximize the data availability by minimizing the data missed rate (MinDmr), based on a file weight and prediction function. Our simulation on the OptorSim shows our MinDmr algorithm achieves better performance overall than others in term of data availability. Results indicate the performance of MinDmr is always better than others with varying prediction functions, job schedulers and file access patterns, as far as the data missing rate is concerned.

This paper presents an extension to a Multi-Level Secure (MLS) data model that requires the classification of data and users into multiple security levels. In MLS systems, cover stories allow information provided to users at lower security levels to differ from information provided to users at higher security levels. Previous versions of the MLS model did not permit cover stories for key attributes, because the key is used to relate the various cover stories for a particular entity. We have extended the MLS model to include non-key related cover stories so that key attributes can also have different values at different security levels. In this paper we describe the necessary model changes and modifications to the relational algebra, which are required to implement non-key related cover stories. We demonstrate the improvements made by these changes and discuss the implementation and performance of a system based on the described concepts.

A mobile computing system consists of mobile and stationary nodes. Checkpointing in mobile systems faces many new challenges, such as low wireless bandwidth, frequent disconnections and lack of stable storage at mobile nodes. Previously proposed checkpoint algorithms for mobile environments do not take into consideration the coordinating message overhead of the mobile participants. In this paper, we propose a new checkpoint algorithm for mobile distributed systems. Our algorithm is nonblocking and minimizes the number of checkpoint participants. We propose a new idea called proxy coordinator. By using proxy coordinator, mobile participants' coordinating message overhead is minimized to O(1), which can otherwise be as high as O(n) in other proposed algorithms.

In a cellular network, the base stations are not necessarily uniformly distributed, and their corresponding cell sizes are not necessarily the same. For example, a cell in a well-populated city cell is usually smaller than a cell in a rural area. To study a cellular network with non-uniform cell sizes, one approach is to use a virtual cellular network with a uniform cell size such that each virtual cell contains at most one base station. This paper has proposed parallel algorithms for meshes with multiple broadcasting to construct virtual mesh and honeycomb cellular networks for non-uniformly distributed base stations. The constructed virtual cellular networks are optimal in the sense that their corresponding uniform cell sizes reach the largest possible. The algorithms run in O(log n) time on a mesh with multiple broadcasting of size n/spl times/n to construct optimal virtual mesh and honeycomb cellular networks for n non-uniformly distributed base stations. Furthermore, those algorithms are time-optimal.

Many real-time database systems, such as military institutions and government agencies, are contained in environments that exhibit restricted access of information, where mandatory access control for security is required. Hence, in addition to timing constraints, real-time database systems have security constraints. Conventional multi-level secure database models are inadequate for time-critical applications and conventional real-time database models do not support security constraints. The objective of this work is to incorporate security constraints in real-time database systems in such a way that not only is security achieved, but achieving security does not degrade real-time performance significantly in terms of deadlines missed. We propose a new optimistic concurrency control algorithm for secure firm real-time databases. Results show that the algorithm performs fairly well in terms of security and timeliness compared to a non-secure algorithm. We argue and show that achieving more security does not necessarily mean more sacrifice in real-time performance.

APPROXIMATE, a query processor that makes approximate answers available if part of the database is unavailable, or if there is not enough time to produce an exact answer, is described. The processor implements approximate query processing, and the accuracy of the approximate result produced improves monotonically with the amount of data retrieved to produce the result. The monotone query processing algorithm of APPROXIMATE works within a standard relational algebra framework. APPROXIMATE maintains semantic information for approximate query processing at an underlying level, and can be implemented on a relational database system with little change to the relational architecture. It is shown how APPROXIMATE is implemented to make effective use of the semantic support. The additional overhead required by APPROXIMATE is described.>

An object-oriented query processor is described that makes approximate answers available if there is not enough time to product an exact answer or if part of the database is unavailable. The accuracy of the approximate result produced improves monotonically with the amount of data retrieved to produce the result. The query-processing algorithm is based on an approximate relational data model and works within a standard relational algebra framework. The query processor maintains an object-oriented view on an underlying level and can be implemented on a relational database system with little change to the relational architecture. It is shown how a monotone query-processing strategy can be implemented, making effective use of semantic information presented by the object-oriented view.>

In real-time database systems, a transaction may not have enough time to complete. In such cases, partial, or imprecise, results can still be produced. The authors have proposed an imprecise result mechanism for producing partial results, which is used to implement timing error recovery in real-time database systems. They also present a model of real-time systems that distinguishes the external data consistency from the internal data consistency maintained by non-real-time systems. Providing a timely response may require sacrificing internal consistency. The authors discuss three examples that have different requirements of data consistency and present algorithms for implementing them.>