Non-Interference is an information flow security property which aims to protect confidential data by ensuring the complete absence of any information flow from high level entities to low level ones. However, this requirement is too demanding when dealing with real applications: indeed, no real policy ever guarantees a total absence of information flow. In order to deal with real applications, it is often necessary to allow mechanisms for downgrading or declassifying information such as information filters and channel control. In this paper we generalize the notion of Persistent Stochastic Non-Interference (PSNI) in order to allow information to flow from a higher to a lower security level through a downgrader. We introduce the notion of Delimited Persistent Stochastic Non-Interference (D_PSNI) and provide two characterizations of it, one expressed in terms of bisimulation-like equivalence checks and another one formulated through unwinding conditions. Then we prove some compositionality properties. Finally, we present a decision algorithm and discuss its complexity.
{"title":"Delimited Persistent Stochastic Non-Interference","authors":"J. Hillston, A. Marin, C. Piazza, S. Rossi","doi":"10.1145/3306309.3306329","DOIUrl":"https://doi.org/10.1145/3306309.3306329","url":null,"abstract":"Non-Interference is an information flow security property which aims to protect confidential data by ensuring the complete absence of any information flow from high level entities to low level ones. However, this requirement is too demanding when dealing with real applications: indeed, no real policy ever guarantees a total absence of information flow. In order to deal with real applications, it is often necessary to allow mechanisms for downgrading or declassifying information such as information filters and channel control. In this paper we generalize the notion of Persistent Stochastic Non-Interference (PSNI) in order to allow information to flow from a higher to a lower security level through a downgrader. We introduce the notion of Delimited Persistent Stochastic Non-Interference (D_PSNI) and provide two characterizations of it, one expressed in terms of bisimulation-like equivalence checks and another one formulated through unwinding conditions. Then we prove some compositionality properties. Finally, we present a decision algorithm and discuss its complexity.","PeriodicalId":113198,"journal":{"name":"Proceedings of the 12th EAI International Conference on Performance Evaluation Methodologies and Tools","volume":"296 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123116302","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Different search engines provide different outputs for the same keyword. This may be due to different definitions of relevance, to different ranking aggregation methods, and/or to different knowledge/anticipation of users' preferences, but rankings are also suspected to be biased towards own content, which may prejudicial to other content providers. In this paper, we make some initial steps toward a rigorous comparison and analysis of search engines, by proposing a definition for a consensual relevance of a page with respect to a keyword, from a set of search engines. More specifically, we look at the results of several search engines for a sample of keywords, and define for each keyword the visibility of a page based on its ranking over all search engines. This allows to define a score of the search engine for a keyword, and then its average score over all keywords. Based on the pages visibility, we can also define the consensus search engine as the one showing the most visible results for each keyword, and discuss how biased results toward specific pages can be highlighted and quantified to provide answers to the search neutrality debate. We have implemented this model and present an analysis of the results.
{"title":"Evaluating the performance and neutrality/bias of search engines","authors":"Ahmed Kamoun, P. Maillé, B. Tuffin","doi":"10.1145/3306309.3306325","DOIUrl":"https://doi.org/10.1145/3306309.3306325","url":null,"abstract":"Different search engines provide different outputs for the same keyword. This may be due to different definitions of relevance, to different ranking aggregation methods, and/or to different knowledge/anticipation of users' preferences, but rankings are also suspected to be biased towards own content, which may prejudicial to other content providers. In this paper, we make some initial steps toward a rigorous comparison and analysis of search engines, by proposing a definition for a consensual relevance of a page with respect to a keyword, from a set of search engines. More specifically, we look at the results of several search engines for a sample of keywords, and define for each keyword the visibility of a page based on its ranking over all search engines. This allows to define a score of the search engine for a keyword, and then its average score over all keywords. Based on the pages visibility, we can also define the consensus search engine as the one showing the most visible results for each keyword, and discuss how biased results toward specific pages can be highlighted and quantified to provide answers to the search neutrality debate. We have implemented this model and present an analysis of the results.","PeriodicalId":113198,"journal":{"name":"Proceedings of the 12th EAI International Conference on Performance Evaluation Methodologies and Tools","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126357351","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Risk-sensitive Markov Decision Process (RSMDP) models are less studied than linear Markov decision models. Linear models optimize only expected cost whereas RSMDP models optimize a combination of expected cost and higher moments of the cost. On the other hand, optimal policies in RSMDP models are generally non-stationary, and need not even be ultimately stationary. This makes optimal policies more difficult to compute and implement in RSMDP models. We provide an algorithm, called Ultimately Stationary Linear Discounted (USLD), to compute an ∈-optimal ultimately stationary policy whose risk-sensitive cost approximates the optimal risk-sensitive cost within any specified degree of accuracy ∈. The algorithm approximates the tail costs with the optimal linear discounted cost, which is then treated as the terminal cost of a finite-horizon RSMDP. For the sake of comparison, we also consider an alternative method from the literature, which we call Ultimately Stationary Tail Off (USTO). USTO is based on the intuition that all decisions beyond a sufficiently large decision epoch make a negligibly small contribution to the total discounted cost. Accordingly, USTO involves optimizing the finite horizon RSMDP cost obtained by ignoring all stage-wise costs that occur after a sufficiently large decision epoch, and then concatenating the resulting finite-horizon policy with an arbitrarily chosen stationary policy to get a ∈-optimal ultimately stationary policy. We provide proofs of risk-sensitive ∈-optimality of policies yielded by USLD and USTO, and compare the performance of both on an inventory control problem.
{"title":"Ultimately Stationary Policies to Approximate Risk-Sensitive Discounted MDPs","authors":"M. UdayKumar, S. Bhat, V. Kavitha, N. Hemachandra","doi":"10.1145/3306309.3306320","DOIUrl":"https://doi.org/10.1145/3306309.3306320","url":null,"abstract":"Risk-sensitive Markov Decision Process (RSMDP) models are less studied than linear Markov decision models. Linear models optimize only expected cost whereas RSMDP models optimize a combination of expected cost and higher moments of the cost. On the other hand, optimal policies in RSMDP models are generally non-stationary, and need not even be ultimately stationary. This makes optimal policies more difficult to compute and implement in RSMDP models. We provide an algorithm, called Ultimately Stationary Linear Discounted (USLD), to compute an ∈-optimal ultimately stationary policy whose risk-sensitive cost approximates the optimal risk-sensitive cost within any specified degree of accuracy ∈. The algorithm approximates the tail costs with the optimal linear discounted cost, which is then treated as the terminal cost of a finite-horizon RSMDP. For the sake of comparison, we also consider an alternative method from the literature, which we call Ultimately Stationary Tail Off (USTO). USTO is based on the intuition that all decisions beyond a sufficiently large decision epoch make a negligibly small contribution to the total discounted cost. Accordingly, USTO involves optimizing the finite horizon RSMDP cost obtained by ignoring all stage-wise costs that occur after a sufficiently large decision epoch, and then concatenating the resulting finite-horizon policy with an arbitrarily chosen stationary policy to get a ∈-optimal ultimately stationary policy. We provide proofs of risk-sensitive ∈-optimality of policies yielded by USLD and USTO, and compare the performance of both on an inventory control problem.","PeriodicalId":113198,"journal":{"name":"Proceedings of the 12th EAI International Conference on Performance Evaluation Methodologies and Tools","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115807343","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Youssef Ait El Mahjoub, J. Fourneau, Hind Castel-Taleb
We study the performance of the optical network architecture NGreen which is based on an optical slotted ring. The random part of the end to end delay comes from the times needed to build an optical container from arriving Data Units and the insertion of the optical container on the ring. Here, to study the first problem, we build a Discrete Time Markov Chain to model the filling of the optical container with Data Units. We take into account a deadline (to have a small latency) and a constraint on a minimal filling of the container (to be energy efficient). We obtain through a numerical analysis using an ad-hoc algorithm we proved, the distribution of the container filling and the distribution of the time needed to build a container. We also compare the algorithm speed with two well known algorithms to compute the steady-state distribution of Markov chains.
{"title":"A numerical approach of the analysis of optical container filling","authors":"Youssef Ait El Mahjoub, J. Fourneau, Hind Castel-Taleb","doi":"10.1145/3306309.3306333","DOIUrl":"https://doi.org/10.1145/3306309.3306333","url":null,"abstract":"We study the performance of the optical network architecture NGreen which is based on an optical slotted ring. The random part of the end to end delay comes from the times needed to build an optical container from arriving Data Units and the insertion of the optical container on the ring. Here, to study the first problem, we build a Discrete Time Markov Chain to model the filling of the optical container with Data Units. We take into account a deadline (to have a small latency) and a constraint on a minimal filling of the container (to be energy efficient). We obtain through a numerical analysis using an ad-hoc algorithm we proved, the distribution of the container filling and the distribution of the time needed to build a container. We also compare the algorithm speed with two well known algorithms to compute the steady-state distribution of Markov chains.","PeriodicalId":113198,"journal":{"name":"Proceedings of the 12th EAI International Conference on Performance Evaluation Methodologies and Tools","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125026351","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The limited performance of a centralized controller can become the bottleneck in a large software defined network. One approach to reduce the flow setup time is to deploy multiple controllers in a network. This increases the number of packets that can be handled per time, but also comes with a communication overhead. The optimization problem in this context is to select the best number of controllers such that the flow setup time can be minimized. In this paper, we set up a queueing model for the response time of a controller. We use the model to evaluate the flow setup time and optimize the optimal number of controllers. We implement a prototype of multiple controllers, measure the response time of the prototype processing different jobs and fit a hyper-Erlang distribution to the response time. We use the fitted distribution in the queueing model to determine the optimal number of the controllers. The queueing analysis shows that the optimal number of controllers decreases with the rate of communication messages among the controllers and increases with the rate of requests from the switches. Our analysis allows us to find best points of operation.
{"title":"Performance Evaluation of the Control Plane in Software Defined Networks","authors":"Zhihao Shang, Han Wu, K. Wolter","doi":"10.1145/3306309.3306336","DOIUrl":"https://doi.org/10.1145/3306309.3306336","url":null,"abstract":"The limited performance of a centralized controller can become the bottleneck in a large software defined network. One approach to reduce the flow setup time is to deploy multiple controllers in a network. This increases the number of packets that can be handled per time, but also comes with a communication overhead. The optimization problem in this context is to select the best number of controllers such that the flow setup time can be minimized. In this paper, we set up a queueing model for the response time of a controller. We use the model to evaluate the flow setup time and optimize the optimal number of controllers. We implement a prototype of multiple controllers, measure the response time of the prototype processing different jobs and fit a hyper-Erlang distribution to the response time. We use the fitted distribution in the queueing model to determine the optimal number of the controllers. The queueing analysis shows that the optimal number of controllers decreases with the rate of communication messages among the controllers and increases with the rate of requests from the switches. Our analysis allows us to find best points of operation.","PeriodicalId":113198,"journal":{"name":"Proceedings of the 12th EAI International Conference on Performance Evaluation Methodologies and Tools","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125193139","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
FemtoClouds are computing platforms, implementing the Fog Computing paradigm, consisting in an ensemble of heterogeneous mobile devices whose users agree to run the tasks offloaded by other users. FemtoClouds are well suited for the execution of Bag-of-Tasks (BoTs) applications, but, being characterized by high resource heterogeneity and volatility, require the availability of scheduling techniques able to effectively deal with ensembles of independently-owned, heterogeneous devices that can suddenly leave the system. In this paper we propose WQR-UD, an online scheduling algorithm that, thanks to the combination of simple task and device selection policies (that do not require any information concerning the applications and the devices) with effective heterogeneity and volatility tolerance mechanisms, is able to effectively schedule a stream of BoT applications on FemtoCloud systems. We assess the ability of WQR-UD to meet its design goals by running an extensive simulation study for a large set of realistic operational scenarios. Our results clearly indicate that WQR-UD is able to effectively schedule a stream of BoT applications on FemtoCloud systems.
{"title":"WQR-UD: An online scheduling algorithm for FemtoClouds","authors":"C. Anglano, M. Canonico, Marco Guazzone","doi":"10.1145/3306309.3306338","DOIUrl":"https://doi.org/10.1145/3306309.3306338","url":null,"abstract":"FemtoClouds are computing platforms, implementing the Fog Computing paradigm, consisting in an ensemble of heterogeneous mobile devices whose users agree to run the tasks offloaded by other users. FemtoClouds are well suited for the execution of Bag-of-Tasks (BoTs) applications, but, being characterized by high resource heterogeneity and volatility, require the availability of scheduling techniques able to effectively deal with ensembles of independently-owned, heterogeneous devices that can suddenly leave the system. In this paper we propose WQR-UD, an online scheduling algorithm that, thanks to the combination of simple task and device selection policies (that do not require any information concerning the applications and the devices) with effective heterogeneity and volatility tolerance mechanisms, is able to effectively schedule a stream of BoT applications on FemtoCloud systems. We assess the ability of WQR-UD to meet its design goals by running an extensive simulation study for a large set of realistic operational scenarios. Our results clearly indicate that WQR-UD is able to effectively schedule a stream of BoT applications on FemtoCloud systems.","PeriodicalId":113198,"journal":{"name":"Proceedings of the 12th EAI International Conference on Performance Evaluation Methodologies and Tools","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133244362","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
There exist many aspects involved in a video turning viral on YouTube. These include properties of the video such as the attractiveness of its title and thumbnail, the recommendation policy of YouTube, marketing and advertising policies and the influence that the video's creator or owner has in social networks. In this work, we study audience retention measures provided by YouTube to video creators which may provide valuable information for improving the videos and for better understanding the viewers' potential interests in them. We then study the question of when is a video too long and can gain from being shortened. We examine consistency between several existing audience retention measures. We end in a proposal for a new audience retention measure and identify its advantages.
{"title":"Measuring Audience Retention in YouTube","authors":"E. Altman, T. Jiménez","doi":"10.1145/3306309.3306322","DOIUrl":"https://doi.org/10.1145/3306309.3306322","url":null,"abstract":"There exist many aspects involved in a video turning viral on YouTube. These include properties of the video such as the attractiveness of its title and thumbnail, the recommendation policy of YouTube, marketing and advertising policies and the influence that the video's creator or owner has in social networks. In this work, we study audience retention measures provided by YouTube to video creators which may provide valuable information for improving the videos and for better understanding the viewers' potential interests in them. We then study the question of when is a video too long and can gain from being shortened. We examine consistency between several existing audience retention measures. We end in a proposal for a new audience retention measure and identify its advantages.","PeriodicalId":113198,"journal":{"name":"Proceedings of the 12th EAI International Conference on Performance Evaluation Methodologies and Tools","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130109064","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper investigates how different battery models perform in terms of battery dynamics and state-of-charge (SOC) estimation. We have selected Shepherd and equivalent circuit models to represent the battery pack of lithium-ion batteries in electric vehicles (EVs). We generate the power demand using existing driving cycles to discharge the battery. We conclude that the EC model captures the dynamics of the battery more accurately than the Shepherd model. Nevertheless, both models produce similar results when we assess the SOC estimation.
{"title":"Battery Models Investigation and Evaluation Using a Power Demand Generated from Driving Cycles","authors":"Saad Alateef, N. Thomas","doi":"10.1145/3306309.3306341","DOIUrl":"https://doi.org/10.1145/3306309.3306341","url":null,"abstract":"This paper investigates how different battery models perform in terms of battery dynamics and state-of-charge (SOC) estimation. We have selected Shepherd and equivalent circuit models to represent the battery pack of lithium-ion batteries in electric vehicles (EVs). We generate the power demand using existing driving cycles to discharge the battery. We conclude that the EC model captures the dynamics of the battery more accurately than the Shepherd model. Nevertheless, both models produce similar results when we assess the SOC estimation.","PeriodicalId":113198,"journal":{"name":"Proceedings of the 12th EAI International Conference on Performance Evaluation Methodologies and Tools","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129178779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
L. Campanile, M. Gribaudo, M. Iacono, Michele Mastroianni
Autonomic routers are the main component on which autonomic networking is founded. Our goal is to provide a first approach performance modeling method that can be usable by networking professionals that are not part of the Performance Evaluation community.
{"title":"Quantitative Modeling of the Behaviour of an Autonomic Router","authors":"L. Campanile, M. Gribaudo, M. Iacono, Michele Mastroianni","doi":"10.1145/3306309.3306344","DOIUrl":"https://doi.org/10.1145/3306309.3306344","url":null,"abstract":"Autonomic routers are the main component on which autonomic networking is founded. Our goal is to provide a first approach performance modeling method that can be usable by networking professionals that are not part of the Performance Evaluation community.","PeriodicalId":113198,"journal":{"name":"Proceedings of the 12th EAI International Conference on Performance Evaluation Methodologies and Tools","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132506621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We study in this paper a multiobjective dynamic programmming where all the criteria are in the form of total expected sum of costs till absorption in some set of states M. We assume that instantaneous costs are strictly positive and make no assumption on the ergodic structure of the Markov Decision Process. Our main result is to extend the linear program solution approach that was previously derived for transient CMDPs (Constrained Markov Decision Processes) to general ergodic structure. Several (additive) cost metrics are defined and (possibly randomized) routing policies are sought which minimize one of the costs subject to constraints over the other objectives.
{"title":"Constrained Markov Decision Processes with Total Expected Cost Criteria","authors":"E. Altman, Said Boularouk, D. Josselin","doi":"10.1145/3306309.3306342","DOIUrl":"https://doi.org/10.1145/3306309.3306342","url":null,"abstract":"We study in this paper a multiobjective dynamic programmming where all the criteria are in the form of total expected sum of costs till absorption in some set of states M. We assume that instantaneous costs are strictly positive and make no assumption on the ergodic structure of the Markov Decision Process. Our main result is to extend the linear program solution approach that was previously derived for transient CMDPs (Constrained Markov Decision Processes) to general ergodic structure. Several (additive) cost metrics are defined and (possibly randomized) routing policies are sought which minimize one of the costs subject to constraints over the other objectives.","PeriodicalId":113198,"journal":{"name":"Proceedings of the 12th EAI International Conference on Performance Evaluation Methodologies and Tools","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126112983","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：