Memory resource balancing for virtualized computing

Virtualization has become a common abstraction layer in modern data centers. By multiplexing hardware resources into multiple virtual machines (VMs) and thus enabling several operating systems to run on the same physical platform simultaneously, it can effectively reduce power consumption and building size or improve security by isolating VMs. In a virtualized system, memory resource management plays a critical role in achieving high resource utilization and performance. Insufficient memory allocation to a VM will degrade its performance dramatically. On the contrary, over-allocation causes waste of memory resources. Meanwhile, a VM’s memory demand may vary significantly. As a result, effective memory resource management calls for a dynamic memory balancer, which, ideally, can adjust memory allocation in a timely manner for each VM based on their current memory demand and thus achieve the best memory utilization and the optimal overall performance. In order to estimate the memory demand of each VM and to arbitrate possible memory resource contention, a widely proposed approach is to construct an LRU-based miss ratio curve (MRC), which provides not only the current working set size (WSS) but also the correlation between performance and the target memory allocation size. Unfortunately, the cost of constructing an MRC is nontrivial. In this dissertation, we first present a low overhead LRU-based memory demand tracking scheme, which includes three orthogonal optimizations: AVL-based LRU organization, dynamic hot set sizing and intermittent memory tracking. Our evaluation results show that, for the whole SPEC CPU 2006 benchmark suite, after applying the three optimizing techniques, the mean overhead of MRC construction is lowered from 173% to only 2%. Based on current WSS, we then predict its trend in the near future and take different strategies for different prediction results. When there is a sufficient amount of physical memory on the host, it locally balances its memory resource for the VMs. Once the local memory resource is insufficient and the memory pressure is predicted to sustain for a sufficiently long time, a relatively expensive solution, VM live migration, is used to move one or more VMs from the hot host to other host(s). Finally, for transient memory pressure, a remote cache is used to alleviate the temporary performance penalty. Our experimental results show that this design achieves 49% center-wide speedup.

Payone : incentive for epidemic protocol-based anonymity system

Anonymity systems maintain the anonymity of communicating nodes by camouflaging them, either with peer nodes generating dummy traffic or with peer nodes participating in the actual communication process. The probability of any adversary breaking down the anonymity of the communicating nodes is inversely proportional to the number of peer nodes participating in the network. Hence to maintain the anonymity of the communicating nodes, a large number of peer nodes are needed. Lack of peer availability weakens the anonymity of any large scale anonymity system. This work proposes PayOne, an incentive based scheme for promoting peer availability. PayOne aims to increase the peer availability by encouraging nodes to participate in the anonymity system by awarding them with incentives and thereby promoting the anonymity strength. Existing incentive schemes are designed for single path based approaches. There is no incentive scheme for multipath based or epidemic based anonymity systems. This work has been specifically designed for epidemic protocols and has been implemented over MuON, one of the latest entries to the area of multicasting based anonymity systems. MuON is a peer-to-peer based anonymity system which uses epidemic protocol for data dissemination. Existing incentive schemes involve paying every intermediate node that is involved in the communication between the initiator and the receiver. These schemes are not appropriate for epidemic based anonymity systems due to the incurred overhead. PayOne differs from the existing schemes because it involves paying a single intermediate node that participates in the network. The intermediate node can be any random node that participates in the communication and does not necessarily need to lie in the communication path between the initiator and the receiver. The light-weight characteristics of PayOne make it viable for large-scale epidemic based anonymity systems.

MODELING AND PROTOCOL DESIGN FOR EMERGING WIRELESS SYSTEMS

Mobile Mesh Network based In-Transit Visibility (MMN-ITV) system facilitates global real-time tracking capability for the logistics system. In-transit containers form a multi-hop mesh network to forward the tracking information to the nearby sinks, which further deliver the information to the remote control center via satellite. The fundamental challenge to the MMN-ITV system is the energy constraint of the battery-operated containers. Coupled with the unique mobility pattern, cross-MMN behavior, and the large-spanned area, it is necessary to investigate the energy-efficient communication of the MMN-ITV system thoroughly. First of all, this dissertation models the energy-efficient routing under the unique pattern of the cross-MMN behavior. A new modeling approach, pseudo-dynamic modeling approach, is proposed to measure the energy-efficiency of the routing methods in the presence of the cross-MMN behavior. With this approach, it could be identified that the shortest-path routing and the load-balanced routing is energy-efficient in mobile networks and static networks respectively. For the MMN-ITV system with both mobile and static MMNs, an energy-efficient routing method, energy-threshold routing, is proposed to achieve the best tradeoff between them. Secondly, due to the cross-MMN behavior, neighbor discovery is executed frequently to help the new containers join the MMN, hence, consumes similar amount of energy as that of the data communication. By exploiting the unique pattern of the cross-MMN behavior, this dissertation proposes energy-efficient neighbor discovery wakeup schedules to save up to 60% of the energy for neighbor discovery. Vehicular Ad Hoc Networks (VANETs)-based inter-vehicle communications is by now growingly believed to enhance traffic safety and transportation management with low cost. The end-to-end delay is critical for the time-sensitive safety applications in VANETs, and can be a decisive performance metric for VANETs. This dissertation presents a complete analytical model to evaluate the end-to-end delay against the transmission range and the packet arrival rate. This model illustrates a significant end-to-end delay increase from non-saturated networks to saturated networks. It hence suggests that the distributed power control and admission control protocols for VANETs should aim at improving the real-time capacity (the maximum packet generation rate without causing saturation), instead of the delay itself. Based on the above model, it could be determined that adopting uniform transmission range for every vehicle may hinder the delay performance improvement, since it does not allow the coexistence of the short path length and the low interference. Clusters are proposed to configure non-uniform transmission range for the vehicles. Analysis and simulation confirm that such configuration can enhance the real-time capacity. In addition, it provides an improved trade off between the end-to-end delay and the network capacity. A distributed clustering protocol with minimum message overhead is proposed, which achieves low convergence time.

MOWER : A NEW DESIGN FOR NON-BLOCKING MISPREDICTION RECOVERY

Mower is a micro-architecture technique which targets branch misprediction penalties in superscalar processors. It speeds-up the misprediction recovery process by dynamically evicting stale instructions and fixing the RAT (Register Alias Table) using explicit branch dependency tracking. Tracking branch dependencies is accomplished by using simple bit matrices. This low-overhead technique allows overlapping of the recovery process with instruction fetching, renaming and scheduling from the correct path. Our evaluation of the mechanism indicates that it yields performance very close to ideal recovery and provides up to 5% speed-up and 2% reduction in power consumption compared to a traditional recovery mechanism using a reorder buffer and a walker. The simplicity of the mechanism should permit easy implementation of Mower in an actual processor.

A FIREWALL MODEL OF FILE SYSTEM SECURITY

File system security is fundamental to the security of UNIX and Linux systems since in these systems almost everything is in the form of a file. To protect the system files and other sensitive user files from unauthorized accesses, certain security schemes are chosen and used by different organizations in their computer systems. A file system security model provides a formal description of a protection system. Each security model is associated with specified security policies which focus on one or more of the security principles: confidentiality, integrity and availability. The security policy is not only about “who” can access an object, but also about “how” a subject can access an object. To enforce the security policies, each access request is checked against the specified policies to decide whether it is allowed or rejected. The current protection schemes in UNIX/Linux systems focus on the access control. Besides the basic access control scheme of the system itself, which includes permission bits, setuid and seteuid mechanism and the root, there are other protection models, such as Capabilities, Domain Type Enforcement (DTE) and Role-Based Access Control (RBAC), supported and used in certain organizations. These models protect the confidentiality of the data directly. The integrity of the data is protected indirectly by only allowing trusted users to operate on the objects. The access control decisions of these models depend on either the identity of the user or the attributes of the process the user can execute, and the attributes of the objects. Adoption of these sophisticated models has been slow; this is likely due to the enormous complexity of specifying controls over a large file system and the need for system administrators to learn a new paradigm for file protection. We propose a new security model: file system firewall. It is an adoption of the familiar network firewall protection model, used to control the data that flows between networked computers, toward file system protection. This model can support decisions of access control based on any system generated attributes about the access requests, e.g., time of day. The access control decisions are not on one entity, such as the account in traditional discretionary access control or the domain name in DTE. In file system firewall, the access decisions are made upon situations on multiple entities. A situation is programmable with predicates on the attributes of subject, object and the system. File system firewall specifies the appropriate actions on these situations. We implemented the prototype of file system firewall on SUSE Linux. Preliminary results of performance tests on the prototype indicate that the runtime overhead is acceptable. We compared file system firewall with TE in SELinux to show that firewall model can accommodate many other access control models. Finally, we show the ease of use of firewall model. When firewall system is restricted to specified part of the system, all the other resources are not affected. This enables a relatively smooth adoption. This fact and that it is a familiar model to system administrators will facilitate adoption and correct use. The user study we conducted on traditional UNIX access control, SELinux and file system firewall confirmed that. The beginner users found it easier to use and faster to learn then traditional UNIX access control scheme and SELinux.

High Temperature Superconductivity: Taming Serendipity

A century after the discovery of superconductivity, the design of new and more useful superconductors remains enigmatic. High-temperature super-conductors offer unique solutions to fundamental grid challenges of the 21st century and hold great promise in addressing global energy challenges in energy production, storage, and distribution. Traditionally guided by serendipity, researchers now endeavor to design new superconductors predictively. Advanced experimental techniques, such as point contact spectroscopy measurements, aid in identifying promising candidates.

Practical, Pluggable Types for a Dynamic Language

Most languages fall into one of two camps: either they adopt a unique, static type system, or they abandon static type-checks for run-time checks. Pluggable types blur this division by (i) making static type systems optional, and (ii) supporting a choice of type systems for reasoning about different kinds of static properties. Dynamic languages can then benefit from static-checking without sacrificing dynamic features or committing to a unique, static type system. But the overhead of adopting pluggable types can be very high, especially if all existing code must be decorated with type annotations before any type-checking can be performed. We propose a practical and pragmatic approach to introduce pluggable type systems to dynamic languages. First of all, only annotated code is type-checked. Second, limited type inference is performed on unannotated code to reduce the number of reported errors. Finally, external annotations can be used to type third-party code. We present Typeplug, a Smalltalk implementation of our framework, and report on experience applying the framework to three different pluggable type systems.

Gravity-based Local Clock Synchronization in Wireless Sensor Networks

Contention-based MAC protocols follow periodic listen/sleep cycles. These protocols face the problem of virtual clustering if different unsynchronized listen/sleep schedules occur in the network, which has been shown to happen in wireless sensor networks. To interconnect these virtual clusters, border nodes maintaining all respective listen/sleep schedules are required. However, this is a waste of energy, if locally a common schedule can be determined. We propose to achieve local synchronization with a mechanism that is similar to gravitation. Clusters represent the mass, whereas synchronization messages sent by each cluster represent the gravitation force of the according cluster. Due to the mutual attraction caused by the clusters, all clusters merge finally. The exchange of synchronization messages itself is not altered by LACAS. Accordingly, LACAS introduces no overhead. Only a not yet used property of synchronization mechanisms is exploited.

Klassifizierung der Methoden eines Ganzheitlichen Produktionssystems unter Berücksichtigung der Anforderungen indirekter Bereiche

Seit den 50er Jahren des Zwanzigsten Jahrhunderts erfährt das Konzept der schlanken Produktion stetig Erweiterungen und Renaissancen [Fau09]. Zielt das Ursprungskonzept noch überwiegend auf eine Effizienzsteigerung im Bereich der Fertigung ab, so werden im Rahmen eines Ganzheitlichen Produktionssystems verstärkt auch die indirekten Bereiche, wie z. B. Konstruktion, Forschung und Entwicklung, Planung oder Vertrieb einbezogen. Im folgenden Beitrag werden die Rahmenbedingungen eines Produktionssystems im indirekten Bereich dargestellt und Empfehlungen hinsichtlich einer strukturellen Erweiterung aktuell bestehender Ordnungsrahmen gegeben.

Konzept eines ultraleichten Elektrohängebahnsystems für innerbetriebliche Beförderung

Die inhaltliche Schwerpunktlegung des Konzepts besteht in Entwurf, Planung und Durchführung eines intelligenten und ultraleichten Elektrohängebahnsystems, welches für leichte Lasten (bis 20 kg) ausgelegt ist und auf einem Seilsystem basiert. Aufgrund dessen soll eine hohe Flexibilität des Systems erreicht werden, sodass dieses jederzeit an die besonderen Gegebenheiten und Aufgabenfelder angepasst werden kann. Es soll insbesondere der Erleichterung der innerbetrieblichen Beförderung dienen. Zielstellung ist die Entwicklung eines Prototyps.

Opportunistic Content-Centric Data Transmission During Short Network Contacts

In this paper, we investigate content-centric data transmission in the context of short opportunistic contacts and base our work on an existing content-centric networking architecture. In case of short interconnection times, file transfers may not be completed and the received information is discarded. Caches in content-centric networks are used for short-term storage and do not guarantee persistence. We implemented a mechanism to extend caching on persistent storage enabling the completion of disrupted content transfers. The mechanisms have been implemented in the CCNx framework and have been evaluated on wireless mesh nodes. Our evaluations using multicast and unicast communication show that the implementation can support content transfers in opportunistic environments without significant processing and storing overhead.

Incremental Dynamic Updates with First-class Contexts

Highly available software systems occasionally need to be updated while avoiding downtime. Dynamic software updates reduce down-time, but still require the system to reach a quiescent state in which a global update can be performed. This can be difficult for multi-threaded systems. We present a novel approach to dynamic updates using first-class contexts, called Theseus. First-class contexts make global updates unnecessary: existing threads run to termination in an old context, while new threads start in a new, updated context; consistency between contexts is ensured with the help of bidirectional transformations. We show that for multi-threaded systems with coherent memory, first-class contexts offer a practical and flexible approach to dynamic updates, with acceptable overhead.

Agent-based Content Retrieval for Opportunistic Content-Centric Networks

In this paper, we describe agent-based content retrieval for opportunistic networks, where requesters can delegate content retrieval to agents, which retrieve the content on their behalf. The approach has been implemented in CCNx, the open source CCN framework, and evaluated on Android smart phones. Evaluations have shown that the overhead of agent delegation is only noticeable for very small content. For content larger than 4MB, agent-based content retrieval can even result in a throughput increase of 20% compared to standard CCN download applications. The requester asks every probe interval for agents that have retrieved the desired content. Evaluations have shown that a probe interval of 30s delivers the best overall performance in our scenario because the number of transmitted notification messages can be decreased by up to 80% without significantly increasing the download time.

Aggregated Multicast — A Comparative Study

Though IP multicast is resource ef£cient in delivering data to a group of members simultaneously, it suffers from scalability problem with the number of concurrently active multicast groups because it requires a router to keep forwarding state for every multicast tree passing through it. To solve this state scalability problem, we proposed a scheme, called aggregated multicast. The key idea is that multiple groups are forced to share a single delivery tree. In our earlier work, we introduced the basic concept of aggregated multicast and presented some initial results to show that multicast state can be reduced. In this paper, we develop a more quantitative assessment of the cost/bene£t trade-offs. We propose an algorithm to assign multicast groups to delivery trees with controllable cost and introduce metrics to measure multicast state and tree management overhead for multicast schemes. We then compare aggregated multicast with conventional multicast schemes, such as source speci£c tree scheme and shared tree scheme. Our extensive simulations show that aggregated multicast can achieve signi£cant routing state and tree management overhead reduction while containing the expense of extra resources (bandwidth waste and tunnelling overhead). We conclude that aggregated multicast is a very cost-effective and promising direction for scalable transit domain multicast provisioning.

Cross-Layer Optimizations for Multimedia Distribution over Wireless Multimedia Sensor Networks and Flying Ad-Hoc Networks with Quality of Experience Support

The proliferation of multimedia content and the demand for new audio or video services have fostered the development of a new era based on multimedia information, which allowed the evolution of Wireless Multimedia Sensor Networks (WMSNs) and also Flying Ad-Hoc Networks (FANETs). In this way, live multimedia services require real-time video transmissions with a low frame loss rate, tolerable end-to-end delay, and jitter to support video dissemination with Quality of Experience (QoE) support. Hence, a key principle in a QoE-aware approach is the transmission of high priority frames (protect them) with a minimum packet loss ratio, as well as network overhead. Moreover, multimedia content must be transmitted from a given source to the destination via intermediate nodes with high reliability in a large scale scenario. The routing service must cope with dynamic topologies caused by node failure or mobility, as well as wireless channel changes, in order to continue to operate despite dynamic topologies during multimedia transmission. Finally, understanding user satisfaction on watching a video sequence is becoming a key requirement for delivery of multimedia content with QoE support. With this goal in mind, solutions involving multimedia transmissions must take into account the video characteristics to improve video quality delivery. The main research contributions of this thesis are driven by the research question how to provide multimedia distribution with high energy-efficiency, reliability, robustness, scalability, and QoE support over wireless ad hoc networks. The thesis addresses several problem domains with contributions on different layers of the communication stack. At the application layer, we introduce a QoE-aware packet redundancy mechanism to reduce the impact of the unreliable and lossy nature of wireless environment to disseminate live multimedia content. At the network layer, we introduce two routing protocols, namely video-aware Multi-hop and multi-path hierarchical routing protocol for Efficient VIdeo transmission for static WMSN scenarios (MEVI), and cross-layer link quality and geographical-aware beaconless OR protocol for multimedia FANET scenarios (XLinGO). Both protocols enable multimedia dissemination with energy-efficiency, reliability and QoE support. This is achieved by combining multiple cross-layer metrics for routing decision in order to establish reliable routes.