The explosion of traffic associated with video content poses significant challenges for mobile content provision. While, on the one hand, mobile video traffic surge is forecast-ed to require significant investments in bandwidth acquisition and infrastructure dimensioning and roll-out, on the other hand, users are not likely to be willing to pay significantly more than today. This increases the pressure to develop solutions capable of making the mobile provision of video more affordable without either affecting user experience or limiting usage. In this respect, this paper proposes a novel methodology for video content delivery which is based on a user video quality perception model. According to this scheme, the video quality of each scene in a movie is selected, from among a finite set of available qualities, with the purpose of reducing the overall bandwidth required to attain a given user experience level targeted by the system for each user and each video. This novel methodology also adopts a clustering approach to identify users with similar Quality of Experience (QoE) profiles and leverages this information for improving the accuracy of user perceived quality predictions. This approach has been validated through a crowd-sourced subjective test evaluation performed with real users using a novel method involving the Amazon Mechanical Turk platform. The results showed that the proposed method is capable of achieving a prediction accuracy in the order of ±0.5 MOS points. This approach can be effectively used to select the video qualities minimizing bandwidth costs while delivering predefined level of perceived quality to the end users.

Abstract—This paper describes a novel QoE-aware adaptive video streaming method that enhances the viewing experience on mobile devices and reduces cellular network bandwidth consumed by Dynamic Adaptive Streaming over HTTP (DASH) by consid-ering perceptual video quality and data rate channel conditions in the bitrate adaptation process. By streaming an optimized video for the particular video quality and channel conditions to a mobile device, we can improve the worst video qualities caused by DASH streaming and reduce quality variations using fewer number of bits. Keywords—QoE, Adaptive video streaming, DASH I.

The explosive growth of Over-the-top (OTT) online video strains capacity of operators' networks, which severely threatens video quality perceived by end users. Since video is very bandwidth consuming, its distribution costs are becoming too high to scale with network investments that are required to support the increasing bandwidth demand. Content providers and operators are searching for solutions to reduce this video traffic load, without degrading their customers' perceived Quality of Experience (QoE). This paper proposes a method that can programmatically optimize video content for desired QoE according to perceptual video quality and device display properties, while achieving bandwidth and storage savings for content providers, operators, and end users. The preliminary results obtained with Samsung Galaxy S3 phone show that up to 60% savings can be achieved by optimizing movies without compromising the perceptible video quality, and up to 70% for perceptible, but not annoying video quality difference. Tailoring video optimization to individual user perception can provide seamless QoE delivery across all users, with a low overhead (i.e., 10%) required to achieve this goal. Finally, two applications of video optimization: QoE-aware delivery and storage, are proposed and examined.

The current content provision methods and associated pricing and business models are challenged by the traffic requirements anticipated for future “data intensive” services. In order to deliver substantially higher peak rates operators will need to deploy a much denser infrastructure and/or acquire more spectrum, thus significantly increasing their CAPEX and OPEX and reducing revenues. To improve the utilization of available network resources this paper presents ActiveCast, a disruptive content delivery paradigm that supports opportunistic content pre-fetching by introducing semantic and context awareness in the currently “agnostic” networking paradigm. The experimental investigations presented in the paper focus on mobile video provision and a content provider, integrated with Facebook and YouTube, has been developed and used to identify socially relevant content for a set of test users. Part of the studies presented in the paper aim at experimentally understanding the structure of the energy costs associated with pre-fetching and on defining a delivery strategy that allows controlling the amount of energy invested. A comparison between a centralized implementation, in which pre-fetching is coordinated by the mobile operators, and an Over-The-Top (OTT) implementation of ActiveCast are also presented. The results show that complementing the context information available at individual user terminals with traffic information, shared by mobile operators through the ActiveCast API, can substantially reduce the energy costs of content delivery, as compared with “on demand” video streaming. Additionally, opportunistically exploiting connections with WiFi APs can amplify the gains already achievable by prefetching on wide area networks.

The arrival of smart phones and tablets, along with a flat rate mobile Internet pricing model have caused increasing adoption of mobile data services. According to recent studies, video has been the main driver of mobile data consumption, having a higher growth rate than any other mobile application. However, streaming a medium/high quality video files can be an issue in a mobile environment where available capacity needs to be shared among a large number of users. Additionally, the energy consumption in mobile devices increases proportionally with the duration of data transfers, which depend on the download data rates achievable by the device. In this respect, adoption of opportunistic content pre-fetching schemes that exploit times and locations with high data rates to deliver content before a user requests it, has the potential to reduce the energy consumption associated with content delivery and improve the user's quality of experience, by allowing playback of pre-stored content with virtually no perceived interruptions or delays. This paper presents a family of opportunistic content pre-fetching schemes and compares their performance to standard on-demand access to content. By adopting a simulation approach on experimental data, collected with monitoring software installed in mobile terminals, we show that content pre-fetching can reduce energy consumption of the mobile devices by up to 30% when compared to the on demand download of the same file, with a time window of 1 hour needed to complete the content prepositioning.

Current wireless networks have no knowledge on the type and characteristics of the specific mobile services and contents they are providing (they are agnostic). On the other hand, associated to richer content and more advanced devices, improvement of Quality of Experience (QoE) levels perceived by each user looks as one of the main goals of the different actors in the telecommunication's ecosystem. In this respect, network awareness of the transferred content types and their quality requirements can be important to deliver them fulfilling user requirements while optimizing the utilization of network resources. In this paper, we explore the effect of including semantic knowledge in the radio resource management (RRM) schemes of cellular systems as a way to reach this balance for video streaming service. To make this possible, we propose a number of semantic-aware RRM schemes using information about the video playback's current status, and investigate their performances based on the reduction of the total time of interruptions (TTI) perceived by users when a video streaming content is played. The achieved results are compared to those of the standard agnostic scheme implemented in the cellular networks. The outcomes show that incorporating semantic-aware RRM schemes, more users can receive the content provided by a network within a tolerable level of TTI, while the number of users receiving High Definition (HD) videos can be increased, which in general terms might reduce the investment in infrastructure to fulfill users' requirements.

The apps paradigm is rapidly changing the way in which content is transferred and consumed in mobile networks. Currently, the traffic loads generated by apps are mostly provided as Over-The-Top (OTT) services, essentially transparent to the cellular network operators, while some of the content is delivered and updated in user terminals through “background transmissions”, without user intervention. With increasing traffic volumes, associated to richer content and more advanced devices, the apps paradigm might create severe system inefficiencies. In this paper we explore a number of apps based content delivery methods and investigate their performances on multiple relevant dimensions, such as: terminal energy consumption, time for accessing in-apps content and their impact on the user experience associated with other mobile data services. The proposed methods include opportunistic content pre-fetching and are characterized by different degrees of context-awareness. One approach considers only context information at the individual user terminals, while another one includes network context information and assumes that content distribution coordination is performed by the network operators. The results show that when one is increasing the numbers of apps, whose content needs to be simultaneously maintained updated, only the operator-driven approach can be feasible, mainly from an energetic perspective. However, for a small number of maintained apps, pre-fetching schemes are superior to standard “on-demand” content delivery solutions, suggesting that pre-fetching should be limited only to the subset of terminal apps with higher user access probabilities.

A novel experimental approach for investigating the performances of context-aware content delivery schemes is presented in this paper. An innovative testbed, capable of remotely controlling multiple terminals, injecting a wide range of traffic loads in real networks and monitoring different performance measures has been developed and utilized for quantifying both the energy costs and user perceived service quality associated with different context- aware content pre-fetching schemes. In the implementation proposed in this paper, the context information required for performing content pre-fetching is extracted and utilized by individual user terminals and does not require any support from mobile operators. The performances of pre-fetching are compared to those of an on-demand content delivery scheme, for both video streaming and file downloading services. The results shows that not only pre-fetching can increase user service appreciation by reducing the time needed to access the information, but it can also significantly lower the amount of energy consumed in user terminals for retrieving the content. Our experiments further indicate that in order to achieve these additional energy gains only limited content prediction capabilities are required, thus making pre-fetching a solid candidate for the provision of a wide range of content types and services in both wide and local area networks.

A context aware model for the delivery of content in mobile networks is presented and studied through simulation. The proposed model is based on predictive knowledge of mobile user behavior related to consumption of content on smart terminals. Content prediction is here exploited to change the time and the location of content delivery, opportunistically utilizing the instantaneously available excess of resources in the network. This is anticipated both to increase network utilization and to enhance user perceived service quality. However, since the accuracy of the context information, and associated content predictions, might have a significant impact on performances, our investigations have been accounting for different content prediction capabilities as well as for various degrees of similarity in users' content requests. The obtained results show that substantial gains both in terms of wireless network efficiency and improved user service perception can be achieved, as compared to classical content delivery methods, for a number of realistic scenarios.

A context aware model for delivery of content in mobile networks is introduced and studied through simulation. The model is based on predictive knowledge of mobile user and mobile group behavior. The predictive behavior is related to consumption of mobile content on smart mobile terminals. The model proposes to change the time and the location of the delivery of predicted content consumption to optimize the wireless network utilization and to improve the user experience. Different content delivery strategies, considering pre-fetching at the user terminals, caching at the base stations and multi-cast wireless transmissions are proposed and investigated. The simulation results show substantial gains in the content delivery efficiency of cellular networks and improved user perceived quality for a number of realistic network operation regimes.

We describe the architecture and the implementation of Multishelf-a decentralized peer-to-peer infomediator. Multishelf was designed and built during a ten-week senior software methodology course using the Proem peer-to-peer platform. We discuss the infomediation problem, present our scalable peer-to-peer solution including the effectiveness of using the Proem peer-to-peer platform, and conclude with plans for future work.

This paper motivates and describes the notion of mobile ad-hoc information systems. Such a system consists of a decentralized and self-organizing network of autonomous, mobile devices that interact as peers. Connectivity is determined by distance between devices; as hosts change their physical location they establish pair-wise communication links based on mutual proximity. We describe application scenarios for mobile ad-hoc information systems and identify technical challenges of a generic software infrastructure. Moreover, we present the goals and architecture of Proem, a peer-to-peer system and development platform for mobile ad-hoc applications. Proem has successfully been used as instructional tool in an advanced software engineering course on peer-to-peer computing.

Presents a formal method to provide temporal equivalence in binary-to-binary translation of real-time applications. We first define three different levels of temporal equivalence, i.e. timing equivalence, invariance and divergence. We then find necessary and sufficient conditions for these three levels of temporal equivalence. We take an interval-based approach to test these conditions. If the generated target program is executable with timing equivalence or invariance, it has to be enforced to do so by inserting synchronization. Synchronization methods that enforce the temporal equivalence are also presented.

A methodology is proposed for performance prediction of an application in a multi-computer architecture. The methodology introduces a two-step approach. In the first step the application (Sphinx I) is considered as a benchmark and its performance in uniprocessor systems is predicted from other standard benchmarks. In the second step the predicted performance values for the compositive modules of the application are used in a partition model to predict the performance of the entire application in a multi-computer architecture. The approach is validated by measuring the actual performance of the application on different two-processor configurations and comparing the results to the predictions made. The methodology for predicting application performance in a multi-computer architecture provides a relatively inexpensive approach to design. Rather than porting the application to the target architecture to measure its performance, the performance can be estimated from standard benchmarks, such as Dhrystone and Linpack. Results from Dhrystone and Linpack benchmarks are readily available, even for new systems, since the results are typically published in advertising literature.>

Performance benchmarks are used to help decide the question: 'which system is faster?' With the increased use of computers in critical systems, there are more and more applications of resources to improve system quality. However, there are no benchmarks that can be used to compare the dependability and robustness of systems in order to answer the question: 'which system is more reliable?' The authors present an attempt at the development of a benchmark to gauge a system's robustness as measured by its ability to tolerate errors. The initial effort produced four primitive benchmark programs. They include file management system, memory access, user application, and C library functions. Each primitive benchmark targets a system functionality and measures its behavior, given erroneous inputs. The authors present the motivation and experimental results for one of these primitive benchmarks in detail followed by an analysis of the results. A methodology is presented to combine the primitive benchmarks to form an overall robustness figure.>

This paper discusses the following: effective instrumentation as the key to effective monitoring; distributed monitoring systems as a basis for general purpose distributed multiprocessors; problems and prospects; performance monitoring, fact and fancy; and software aspects.>

The MACS (multiple active context system) architecture is a high-performance, multiple context, cacheless processor designed around the goal of predictability. Multiple contexts provide highly predictable memory and pipeline performance, while at the same time allowing multiple threads to execute with independent timing characteristics. The design emphasis is on provable execution speed and independent timing of each context. Task-level parallelism is used to maintain high processor throughput while individual threads execute at a relatively slow, but very predictable, rate. The authors investigate the suitability of such a system to real time and determine how predictability varies as a function of parallelism (performance), memory issue latency and memory size.>

An approach for efficiently mapping parallel applications onto parallel MIMD machine architectures is introduced. The applicability of this approach to uniform memory-access multiprocessors is demonstrated. It is shown that an intermediate layer of abstraction between the application level and the parallel architecture level is conducive not only to a better software productivity but also to performance efficient programs. The intermediate layer consists of a set of commonly used parallel programming paradigms (implementation machines). A mathematical representation and a pragmatic representation are provided for each implementation machine (IM). The user maps the application onto one or a set of IMs and the system implements the IMs efficiently on the underlying parallel machine.>

The results of several experiments conducted using the fault-injection-based automated testing (FIAT) system are presented. FIAT is capable of emulating a variety of distributed system architectures, and it provides the capabilities to monitor system behavior and inject faults for the purpose of experimental characterization and validation of a system's dependability. The experiments consists of exhaustively injecting three separate fault types into various locations, encompassing both the code and data portions of memory images, of two distinct applications executed with several different data values and sizes. Fault types are variations of memory bit faults. The results show that there are a limited number of system-level fault manifestations. These manifestations follow a normal distribution for each fault type. Error detection latencies are found to be normally distributed. The methodology can be used to predict the system-level fault responses during the system design stage.>

A model for predicting multiprocessor performance on iterative algorithms is developed. Each iteration consists of some amount of access to global data and some amount of local processing. The iterations may be synchronous or asynchronous, and the processors may or may not incur waiting time, depending on the relationship between the access time and processing time. The effect on performance of the speed of the processor, memory, and the interconnection network is studied. The model also illustrates the significant impact on performance of decomposing an algorithm into parallel processes. The model's predictions are calibrated with experimental measurements.>

article The influence of parallel decomposition strategies on the performance of multiprocessor systems Share on Authors: Dalibor Vrsalovic Department of Computer Science Carnegie-Mellon University Department of Computer Science Carnegie-Mellon UniversityView Profile , Edward F. Gehringer Department of Computer Science Carnegie-Mellon University Department of Computer Science Carnegie-Mellon UniversityView Profile , Zary Z. Segall Department of Computer Science Carnegie-Mellon University Department of Computer Science Carnegie-Mellon UniversityView Profile , Daniel P. Siewiorek Department of Computer Science Carnegie-Mellon University Department of Computer Science Carnegie-Mellon UniversityView Profile Authors Info & Claims ACM SIGARCH Computer Architecture NewsVolume 13Issue 3June 1985 pp 396–405https://doi.org/10.1145/327070.327372Online:01 June 1985Publication History 26citation241DownloadsMetricsTotal Citations26Total Downloads241Last 12 Months1Last 6 weeks0 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access

This paper presents two cache schemes for a shared-memory shared bus multiprocessor. Both schemes feature decentralized consistency control and dynamic type classification of the datum cached (i.e. read-only, local, or shared). It is shown how to exploit these features to minimize the shared bus traffic. The broadcasting ability of the shared bus is used not only to signal an event but also to distribute data. In addition, by introducing a new synchronization construct, i.e. the Test-and-Test-and-Set instruction, many of the traditional. parallell processing “hot spots” or bottlenecks are eliminated. Sketches of formal correctness proofs for the proposed schemes are also presented. It appears that moderately large parallel processors can be designed by employing the principles presented in this paper.

This paper introduces the concept of an integrated instrumentation environment (IIE) for multiprocessors. The primary objective of such an environment is to assist the user in the process of experimentation. The emphasis in an IIE is on experiment management (including stimulus generation, monitoring, data collection and analysis), rather than on techniques for program development as in conventional programming environments. We believe the functionality of the two environments should eventually be provided in one comprehensive environment.