In modern hospitals, the role of Clinical Decision Support Systems (CDSS) in assisting care providers is well-established. Most conventional CDSS systems are built on the availability of patient data in the form of structured Electronic Health Records. However, a significant percentage of patient data is still stored in the form of unstructured clinical text notes, especially in developing countries. These contain abundant patient-specific information, which has so far remained largely under-utilized in powering CDSS applications. In this paper, we attempt to build one such CDSS system for patient mortality prediction, using unstructured clinical records. Effectiveness of such prediction models largely depends on optimally capturing latent concept features, thus, word representation quality is of utmost importance. We experiment with three popular word embedding models - Word2Vec, FastText and GloVe for generating word embeddings of unstructured nursing notes of patients from a standard, open dataset, MIMIC-III. These word representations are used as features to train machine learning classifiers to build ICU mortality prediction models, a critical CDSS in ICUs of hospitals. Experimental validation showed that a model built on Word2Vec Skipgram based Random Forest classifier was the most optimal word embedding based mortality prediction model, that outperformed traditional severity scores like SAPS-II, SOFA, APS-III and OASIS, by a significant margin of 43-52%.
{"title":"Evaluating the quality of word representation models for unstructured clinical Text based ICU mortality prediction","authors":"G. Krishnan, Sowmya S Kamath","doi":"10.1145/3288599.3297118","DOIUrl":"https://doi.org/10.1145/3288599.3297118","url":null,"abstract":"In modern hospitals, the role of Clinical Decision Support Systems (CDSS) in assisting care providers is well-established. Most conventional CDSS systems are built on the availability of patient data in the form of structured Electronic Health Records. However, a significant percentage of patient data is still stored in the form of unstructured clinical text notes, especially in developing countries. These contain abundant patient-specific information, which has so far remained largely under-utilized in powering CDSS applications. In this paper, we attempt to build one such CDSS system for patient mortality prediction, using unstructured clinical records. Effectiveness of such prediction models largely depends on optimally capturing latent concept features, thus, word representation quality is of utmost importance. We experiment with three popular word embedding models - Word2Vec, FastText and GloVe for generating word embeddings of unstructured nursing notes of patients from a standard, open dataset, MIMIC-III. These word representations are used as features to train machine learning classifiers to build ICU mortality prediction models, a critical CDSS in ICUs of hospitals. Experimental validation showed that a model built on Word2Vec Skipgram based Random Forest classifier was the most optimal word embedding based mortality prediction model, that outperformed traditional severity scores like SAPS-II, SOFA, APS-III and OASIS, by a significant margin of 43-52%.","PeriodicalId":346177,"journal":{"name":"Proceedings of the 20th International Conference on Distributed Computing and Networking","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129623678","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}
While executing a client application, comprising a sequence of one or more storage operations (read, write, etc.), replicated datastores provide the option of tuning per-operation client-centric consistency settings, Weaker consistency settings allow client applications to execute with lower observed latency due to less coordination among the replica, improving throughput. To enable a client application running on a geo-replicated datastore to meet stringent SLA (Service Level Agreement) deadlines, i.e., latency thresholds provided in the SLA, users apply weak client-centric consistency settings. Strong consistency settings may result in higher latency due to coordination among greater number of replicas. However, since a weak consistency setting requires only a partial quorum of replicas to coordinate, some of the replicas may be overwritten by conflicting updates from concurrent clients, which, in turn, may result in violation of correctness conditions specified by the developer. Such correctness conditions are provided as postconditions that impose restrictions on the values observed by the clients. Enforcing correctness conditions under weak consistency settings on geo-replicated datastores from the client application code is a tedious and challenging task. The task is even more complicated when considering application-level SLA deadlines. Deadline (latency) is one of the most important SLA parameters for optimizing client-centric performance. To this end, we present Consistify, a novel framework, that acts as an interface layer between client applications and underlying data-stores, and enables client applications to execute with weak per-operation consistency settings, while simultaneously respecting: 1) the correctness conditions, specified using logical predicates, and 2) the SLA deadline. Using benchmark workloads, we experimentally demonstrate that Consistify outperforms the state-of-the-art systems, namely QUELA. For several other systems that provide flexible consistency settings, namely Bolt-on causal system, SIEVE, and Cure, we provide a comparison of the throughput and latency observed with Consistify against the numbers reported with those systems in prior papers.
{"title":"Consistify: preserving correctness and SLA under weak consistency","authors":"Subhajit Sidhanta, S. Mukhopadhyay, W. Golab","doi":"10.1145/3288599.3288630","DOIUrl":"https://doi.org/10.1145/3288599.3288630","url":null,"abstract":"While executing a client application, comprising a sequence of one or more storage operations (read, write, etc.), replicated datastores provide the option of tuning per-operation client-centric consistency settings, Weaker consistency settings allow client applications to execute with lower observed latency due to less coordination among the replica, improving throughput. To enable a client application running on a geo-replicated datastore to meet stringent SLA (Service Level Agreement) deadlines, i.e., latency thresholds provided in the SLA, users apply weak client-centric consistency settings. Strong consistency settings may result in higher latency due to coordination among greater number of replicas. However, since a weak consistency setting requires only a partial quorum of replicas to coordinate, some of the replicas may be overwritten by conflicting updates from concurrent clients, which, in turn, may result in violation of correctness conditions specified by the developer. Such correctness conditions are provided as postconditions that impose restrictions on the values observed by the clients. Enforcing correctness conditions under weak consistency settings on geo-replicated datastores from the client application code is a tedious and challenging task. The task is even more complicated when considering application-level SLA deadlines. Deadline (latency) is one of the most important SLA parameters for optimizing client-centric performance. To this end, we present Consistify, a novel framework, that acts as an interface layer between client applications and underlying data-stores, and enables client applications to execute with weak per-operation consistency settings, while simultaneously respecting: 1) the correctness conditions, specified using logical predicates, and 2) the SLA deadline. Using benchmark workloads, we experimentally demonstrate that Consistify outperforms the state-of-the-art systems, namely QUELA. For several other systems that provide flexible consistency settings, namely Bolt-on causal system, SIEVE, and Cure, we provide a comparison of the throughput and latency observed with Consistify against the numbers reported with those systems in prior papers.","PeriodicalId":346177,"journal":{"name":"Proceedings of the 20th International Conference on Distributed Computing and Networking","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129856844","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 the problem of dynamically selecting sensors to maximize the number of steps covered on the path of a mobile target. The sensors are selected from those already deployed in a network as the target moves. The total number of sensors selected and rounds of selection are limited according to budget constraints. The variables are which sensors to select, when, and how. The best settings depend on how much is known about the target's path, such as its start and end locations and mobility pattern, which is updated as the target moves. We study how coverage is affected by these parameters, and how much such dynamic selection strategies can improve coverage over sensors selected in advance.
{"title":"Dynamic sensor selection for path coverage","authors":"S. Shamoun, T. Abdelzaher, A. Bar-Noy","doi":"10.1145/3288599.3288627","DOIUrl":"https://doi.org/10.1145/3288599.3288627","url":null,"abstract":"We study the problem of dynamically selecting sensors to maximize the number of steps covered on the path of a mobile target. The sensors are selected from those already deployed in a network as the target moves. The total number of sensors selected and rounds of selection are limited according to budget constraints. The variables are which sensors to select, when, and how. The best settings depend on how much is known about the target's path, such as its start and end locations and mobility pattern, which is updated as the target moves. We study how coverage is affected by these parameters, and how much such dynamic selection strategies can improve coverage over sensors selected in advance.","PeriodicalId":346177,"journal":{"name":"Proceedings of the 20th International Conference on Distributed Computing and Networking","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128369365","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}
Delay tolerant network (DTN) is successfully proposed for setting up emergency post disaster communication networks when normal communication infrastructure is incapacitated. Performance of such networks get affected by selfish nodes that do not participate in message forwarding. Thus, nodes must receive satisfactory rewards for cooperation. The available incentive schemes either rely on central trusted authorities or do not use an explicit and secure digital currency. Blockchain, a decentralized digital ledger of immutable transactions, is an attractive approach for addressing the incentive challenges in peer-to-peer networks that lack central trusted authorities. Bitcoin, the Blockchain based cryptocurrency, make it possible to devise practical credit based incentive schemes for such networks. In this paper, we propose a Blockchain based incentive scheme for DTN based post disaster communication network that uses Bitcoin to incentivize nodes for cooperation. The scheme uses a novel reward strategy to bring rationality in the incentivizing process.
{"title":"A blockchain based incentive scheme for post disaster opportunistic communication over DTN","authors":"Chandrima Chakrabarti, Souvik Basu","doi":"10.1145/3288599.3295584","DOIUrl":"https://doi.org/10.1145/3288599.3295584","url":null,"abstract":"Delay tolerant network (DTN) is successfully proposed for setting up emergency post disaster communication networks when normal communication infrastructure is incapacitated. Performance of such networks get affected by selfish nodes that do not participate in message forwarding. Thus, nodes must receive satisfactory rewards for cooperation. The available incentive schemes either rely on central trusted authorities or do not use an explicit and secure digital currency. Blockchain, a decentralized digital ledger of immutable transactions, is an attractive approach for addressing the incentive challenges in peer-to-peer networks that lack central trusted authorities. Bitcoin, the Blockchain based cryptocurrency, make it possible to devise practical credit based incentive schemes for such networks. In this paper, we propose a Blockchain based incentive scheme for DTN based post disaster communication network that uses Bitcoin to incentivize nodes for cooperation. The scheme uses a novel reward strategy to bring rationality in the incentivizing process.","PeriodicalId":346177,"journal":{"name":"Proceedings of the 20th International Conference on Distributed Computing and Networking","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122102348","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}
Divya Saxena, Suyash Mahar, V. Raychoudhury, Jiannong Cao
Named Data Networking (NDN) is the most promising candidate among the proposed content-based future Internet architectures. In NDN, Forwarding Information Base (FIB) maintains name prefixes and their corresponding outgoing interface(s) and forwards incoming packets by calculating the longest prefix match (LPM) of their content names (CNs). A CN in NDN is of variable-length and is maintained using a hierarchical structure. Therefore, to perform name lookup for packet forwarding at wire speed is a challenging task. However, the use of GPUs can lead to much better lookup speeds than CPU but, they are often limited by the CPU-GPU transfer latencies. In this paper, we exploit the massive parallel processing power of FPGA technology and propose a scalable, high-speed on-chip SRAM-based NDN name forwarding scheme for FIB (OnChip-FIB) using Field-Programmable Gate Arrays (FPGA). OnChip-FIB scales well as the number of prefixes grow, due to low storage complexity and low resource utilization. Extensive simulation results show that the OnChip-FIB scheme can achieve 1.06 μs measured lookup latency with a 26% on-chip block memory usage in a single Xilinx UltraScale FPGA for 50K named dataset.
{"title":"Scalable, high-speed on-chip-based NDN name forwarding using FPGA","authors":"Divya Saxena, Suyash Mahar, V. Raychoudhury, Jiannong Cao","doi":"10.1145/3288599.3288613","DOIUrl":"https://doi.org/10.1145/3288599.3288613","url":null,"abstract":"Named Data Networking (NDN) is the most promising candidate among the proposed content-based future Internet architectures. In NDN, Forwarding Information Base (FIB) maintains name prefixes and their corresponding outgoing interface(s) and forwards incoming packets by calculating the longest prefix match (LPM) of their content names (CNs). A CN in NDN is of variable-length and is maintained using a hierarchical structure. Therefore, to perform name lookup for packet forwarding at wire speed is a challenging task. However, the use of GPUs can lead to much better lookup speeds than CPU but, they are often limited by the CPU-GPU transfer latencies. In this paper, we exploit the massive parallel processing power of FPGA technology and propose a scalable, high-speed on-chip SRAM-based NDN name forwarding scheme for FIB (OnChip-FIB) using Field-Programmable Gate Arrays (FPGA). OnChip-FIB scales well as the number of prefixes grow, due to low storage complexity and low resource utilization. Extensive simulation results show that the OnChip-FIB scheme can achieve 1.06 μs measured lookup latency with a 26% on-chip block memory usage in a single Xilinx UltraScale FPGA for 50K named dataset.","PeriodicalId":346177,"journal":{"name":"Proceedings of the 20th International Conference on Distributed Computing and Networking","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130176948","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}
Etienne Mauffret, D. Jeanneau, L. Arantes, Pierre Sens
Mutual exclusion is one of the fundamental problems in distributed computing but existing mutual exclusion algorithms are unadapted to the dynamics and lack of membership knowledge of current distributed systems (e.g., mobile ad-hoc networks, peer-to-peer systems, etc.). Additionally, in order to circumvent the impossibility of solving mutual exclusion in asynchronous message passing systems where processes can crash, some solutions include the use of (T + Σl) [3], which is the weakest failure detector to solve mutual exclusion in known static distributed systems. In this paper, we define a new failure detector TΣlr which is equivalent to (T + Σl) in known static systems, and prove that T Σlr is the weakest failure detector to solve mutual exclusion in unknown dynamic systems with partial memory losses. We consider that crashed processes may recover.
{"title":"The weakest failure detector to solve the mutual exclusion problem in an unknown dynamic environment","authors":"Etienne Mauffret, D. Jeanneau, L. Arantes, Pierre Sens","doi":"10.1145/3288599.3288608","DOIUrl":"https://doi.org/10.1145/3288599.3288608","url":null,"abstract":"Mutual exclusion is one of the fundamental problems in distributed computing but existing mutual exclusion algorithms are unadapted to the dynamics and lack of membership knowledge of current distributed systems (e.g., mobile ad-hoc networks, peer-to-peer systems, etc.). Additionally, in order to circumvent the impossibility of solving mutual exclusion in asynchronous message passing systems where processes can crash, some solutions include the use of (T + Σl) [3], which is the weakest failure detector to solve mutual exclusion in known static distributed systems. In this paper, we define a new failure detector TΣlr which is equivalent to (T + Σl) in known static systems, and prove that T Σlr is the weakest failure detector to solve mutual exclusion in unknown dynamic systems with partial memory losses. We consider that crashed processes may recover.","PeriodicalId":346177,"journal":{"name":"Proceedings of the 20th International Conference on Distributed Computing and Networking","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134050341","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}
P. Paul, Naman Mehta, Aman Kumar Das, Sourav Agarwal, Sujoy Saha, S. Nandi
Dissemination of information after a large-scale disaster is always a difficult task and hinders the possibilities of proper decision making. Relief work and rescuing survivors take a major hit without a proper map showcasing the current scenario of the disaster-hit terrain. All due to the lack of traditional communication channels after a disaster. However, we want to change this with the proposition of Android-based mobile application SURAKSHIT that prepares 'localized' crisis map through 'offline' crowdsourcing of circumstantial data and distribution of the same in a smooth manner using a delay-tolerant network. The information gap between the first responders and local survivors can thus be greatly reduced with proper knowledge about the affected area collected by the volunteers. Multimedia data accumulated with the application can further assist the rescue forces in planning rescue-relief operations better.
{"title":"SURAKSHIT: a smartphone-based application for 'localized' GIS data aggregation in absence of internet","authors":"P. Paul, Naman Mehta, Aman Kumar Das, Sourav Agarwal, Sujoy Saha, S. Nandi","doi":"10.1145/3288599.3295588","DOIUrl":"https://doi.org/10.1145/3288599.3295588","url":null,"abstract":"Dissemination of information after a large-scale disaster is always a difficult task and hinders the possibilities of proper decision making. Relief work and rescuing survivors take a major hit without a proper map showcasing the current scenario of the disaster-hit terrain. All due to the lack of traditional communication channels after a disaster. However, we want to change this with the proposition of Android-based mobile application SURAKSHIT that prepares 'localized' crisis map through 'offline' crowdsourcing of circumstantial data and distribution of the same in a smooth manner using a delay-tolerant network. The information gap between the first responders and local survivors can thus be greatly reduced with proper knowledge about the affected area collected by the volunteers. Multimedia data accumulated with the application can further assist the rescue forces in planning rescue-relief operations better.","PeriodicalId":346177,"journal":{"name":"Proceedings of the 20th International Conference on Distributed Computing and Networking","volume":"252 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121240556","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}
Given a trace of a distributed computation and a desired predicate, the predicate detection problem is to find a consistent global state that satisfies the given predicate. The predicate detection problem has many applications in the testing and runtime verification of parallel and distributed systems. We show that many problems related to predicate detection are in the parallel complexity class NC, the set of decision problems decidable in polylogarithmic time on a parallel computer with a polynomial number of processors. Given a computation on n processes with at most m local states per process, our parallel algorithm to detect a given conjunctive predicate takes O(log mn) time and O(m3n3 log mn) work. The sequential algorithm takes O(mn2) time. For data race detection, we give a parallel algorithm that takes O(log mn log n) time, also placing that problem in NC. This is the first work, to the best of our knowledge, that places the parallel complexity of such predicate detection problems in the class NC.
{"title":"Parallel algorithms for predicate detection","authors":"V. Garg, Rohan Garg","doi":"10.1145/3288599.3288604","DOIUrl":"https://doi.org/10.1145/3288599.3288604","url":null,"abstract":"Given a trace of a distributed computation and a desired predicate, the predicate detection problem is to find a consistent global state that satisfies the given predicate. The predicate detection problem has many applications in the testing and runtime verification of parallel and distributed systems. We show that many problems related to predicate detection are in the parallel complexity class NC, the set of decision problems decidable in polylogarithmic time on a parallel computer with a polynomial number of processors. Given a computation on n processes with at most m local states per process, our parallel algorithm to detect a given conjunctive predicate takes O(log mn) time and O(m3n3 log mn) work. The sequential algorithm takes O(mn2) time. For data race detection, we give a parallel algorithm that takes O(log mn log n) time, also placing that problem in NC. This is the first work, to the best of our knowledge, that places the parallel complexity of such predicate detection problems in the class NC.","PeriodicalId":346177,"journal":{"name":"Proceedings of the 20th International Conference on Distributed Computing and Networking","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123388030","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}
Patrícia R. Sousa, André Cirne, João S. Resende, Rolando Martins, L. Antunes
The number of devices connected to the Internet has been increasing exponentially. There is a substantial amount of data being exchanged among numerous connected devices. The added convenience brought by these devices spans across multiple facets of everyday life, such as drivers reporting an accident through dash cams, patients monitoring their own health, and companies controlling the safety of their facilities. However, it is critical to increase safety and privacy across the data generated and propagated by these devices. Previous works have focused mainly on device management and relied on centralized solutions namely Public Key Infrastructure (PKI). This paper describes a novel mechanism that ensures secure autonomous communication between Internet of Things (IoT) devices, while using a completely decentralized solution that mitigates the classical single points-of-failure problem. This is accomplished by a new peer-to-peer protocol using Short Authentication Strings (SAS), in which verification is made through a Limited-Location Channel (LLC).
{"title":"pTASC: trustable autonomous secure communications","authors":"Patrícia R. Sousa, André Cirne, João S. Resende, Rolando Martins, L. Antunes","doi":"10.1145/3288599.3288623","DOIUrl":"https://doi.org/10.1145/3288599.3288623","url":null,"abstract":"The number of devices connected to the Internet has been increasing exponentially. There is a substantial amount of data being exchanged among numerous connected devices. The added convenience brought by these devices spans across multiple facets of everyday life, such as drivers reporting an accident through dash cams, patients monitoring their own health, and companies controlling the safety of their facilities. However, it is critical to increase safety and privacy across the data generated and propagated by these devices. Previous works have focused mainly on device management and relied on centralized solutions namely Public Key Infrastructure (PKI). This paper describes a novel mechanism that ensures secure autonomous communication between Internet of Things (IoT) devices, while using a completely decentralized solution that mitigates the classical single points-of-failure problem. This is accomplished by a new peer-to-peer protocol using Short Authentication Strings (SAS), in which verification is made through a Limited-Location Channel (LLC).","PeriodicalId":346177,"journal":{"name":"Proceedings of the 20th International Conference on Distributed Computing and Networking","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122452022","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}
F. Bronzino, S. Maheshwari, I. Seskar, D. Raychaudhuri
This paper presents NOVN: a novel network virtualization framework aimed at providing an efficient and low overhead solution for deploying Virtual Networks (VNs) based on the concept of named-objects. Name based communication paradigms, realized by separating object names and network addresses through a logically centralized globally distributed Name Resolution Service (NRS), can be used to build a natural and efficient architecture for virtual networks. The proposed VN framework exploits the name based abstraction to create customized networking for distributed services which benefit from an awareness of network topology and routing. A specific example of edge cloud computing is presented in which VN's are used to realize "application specific routing" (ASR) for efficiently connecting users with cloud resources. Experimental results are presented for validation of the proposed VN architecture using a software router implementation running on the ORBIT testbed. The results validate the feasibility of the named-object approach, showing minimal VN processing, control overhead, and latency. The results also validate application aware ASR routing functionality for an example latency constrained edge cloud service scenario.
{"title":"NOVN: named-object based virtual network architecture","authors":"F. Bronzino, S. Maheshwari, I. Seskar, D. Raychaudhuri","doi":"10.1145/3288599.3288637","DOIUrl":"https://doi.org/10.1145/3288599.3288637","url":null,"abstract":"This paper presents NOVN: a novel network virtualization framework aimed at providing an efficient and low overhead solution for deploying Virtual Networks (VNs) based on the concept of named-objects. Name based communication paradigms, realized by separating object names and network addresses through a logically centralized globally distributed Name Resolution Service (NRS), can be used to build a natural and efficient architecture for virtual networks. The proposed VN framework exploits the name based abstraction to create customized networking for distributed services which benefit from an awareness of network topology and routing. A specific example of edge cloud computing is presented in which VN's are used to realize \"application specific routing\" (ASR) for efficiently connecting users with cloud resources. Experimental results are presented for validation of the proposed VN architecture using a software router implementation running on the ORBIT testbed. The results validate the feasibility of the named-object approach, showing minimal VN processing, control overhead, and latency. The results also validate application aware ASR routing functionality for an example latency constrained edge cloud service scenario.","PeriodicalId":346177,"journal":{"name":"Proceedings of the 20th International Conference on Distributed Computing and Networking","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124707204","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：