The emerging publish/subscribe communication paradigm for building large-scale distributed event notification systems, has been shown to exhibit excellent performance and scalability characteristics. Moreover, some work also focus on providing reliability and availability guarantees in the face of node crash and link failures. Such publish/subscribe systems are commonly used in cloud computing infrastructures. However, addressing the dependability concern due to malicious attacks or unintentional software errors, which can potentially corrupt the system, has largely been left untouched by researchers. In this paper, we first identify some of the potential problem areas related to Byzantine behavior in the publish/subscribe paradigm. Secondly, we propose several directions of research for designing a Byzantine fault-tolerant publish/subscribe system suitable for use as a cloud computing infrastructure.
{"title":"Byzantine Fault-Tolerant Publish/Subscribe: A Cloud Computing Infrastructure","authors":"Tiancheng Chang, H. Meling","doi":"10.1109/SRDS.2012.14","DOIUrl":"https://doi.org/10.1109/SRDS.2012.14","url":null,"abstract":"The emerging publish/subscribe communication paradigm for building large-scale distributed event notification systems, has been shown to exhibit excellent performance and scalability characteristics. Moreover, some work also focus on providing reliability and availability guarantees in the face of node crash and link failures. Such publish/subscribe systems are commonly used in cloud computing infrastructures. However, addressing the dependability concern due to malicious attacks or unintentional software errors, which can potentially corrupt the system, has largely been left untouched by researchers. In this paper, we first identify some of the potential problem areas related to Byzantine behavior in the publish/subscribe paradigm. Secondly, we propose several directions of research for designing a Byzantine fault-tolerant publish/subscribe system suitable for use as a cloud computing infrastructure.","PeriodicalId":447700,"journal":{"name":"2012 IEEE 31st Symposium on Reliable Distributed Systems","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116878867","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 a novel efficient approach to utilize multiple radio interfaces for enhancing the performance of reliable multicasts from a single sender to a group of receivers. In the proposed scheme, one radio channel (and interface) is dedicated only for recovery information transmissions. We apply this concept to both ARQ and hybrid ARQ+FEC protocols, formally analyzing the number of packets each receiver needs to process in both our approach and in the common single channel approach. We also present a corresponding efficient protocol, and study its performance by simulation. Both the formal analysis and the simulations demonstrate the benefits of our scheme.
{"title":"Efficient and Reliable Multicast in Multi-radio Networks","authors":"R. Friedman, Alex Kogan","doi":"10.1109/SRDS.2012.22","DOIUrl":"https://doi.org/10.1109/SRDS.2012.22","url":null,"abstract":"This paper investigates a novel efficient approach to utilize multiple radio interfaces for enhancing the performance of reliable multicasts from a single sender to a group of receivers. In the proposed scheme, one radio channel (and interface) is dedicated only for recovery information transmissions. We apply this concept to both ARQ and hybrid ARQ+FEC protocols, formally analyzing the number of packets each receiver needs to process in both our approach and in the common single channel approach. We also present a corresponding efficient protocol, and study its performance by simulation. Both the formal analysis and the simulations demonstrate the benefits of our scheme.","PeriodicalId":447700,"journal":{"name":"2012 IEEE 31st Symposium on Reliable Distributed Systems","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116729924","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}
Cloud computing enables elastic and dynamic resource provisioning while providing cost-effective computing solutions. However, while cloud computing provides customers access to scalable and elastic resources, it does not guarantee the user's expectations of Quality of Service (QoS). This is because a number of customers share resources in the cloud infrastructure simultaneously: compute-intensive processes and network traffic associated with one customer often impact the performance of other applications operated on the same infrastructure in unexpected ways. The inability of the cloud to enforce QoS and provide execution guarantees prevents cloud computing from becoming useful for distributed, real-time and embedded (DRE) systems. Providing the required levels of service to support DRE systems in the cloud is complicated for a variety of reasons: (1) lack of effective monitoring that prevents timely auto-scaling needed for DRE systems, (2) hyper visors and data-center networks that do not support real-time scheduling of resources, and (3) absence of efficient and predictable fault tolerant mechanisms with acceptable overhead and consistency. This paper describes ongoing and proposed doctoral research to address these challenges.
{"title":"Strategies for Reliable, Cloud-Based Distributed Real-Time and Embedded Systems","authors":"Kyoungho An","doi":"10.1109/SRDS.2012.69","DOIUrl":"https://doi.org/10.1109/SRDS.2012.69","url":null,"abstract":"Cloud computing enables elastic and dynamic resource provisioning while providing cost-effective computing solutions. However, while cloud computing provides customers access to scalable and elastic resources, it does not guarantee the user's expectations of Quality of Service (QoS). This is because a number of customers share resources in the cloud infrastructure simultaneously: compute-intensive processes and network traffic associated with one customer often impact the performance of other applications operated on the same infrastructure in unexpected ways. The inability of the cloud to enforce QoS and provide execution guarantees prevents cloud computing from becoming useful for distributed, real-time and embedded (DRE) systems. Providing the required levels of service to support DRE systems in the cloud is complicated for a variety of reasons: (1) lack of effective monitoring that prevents timely auto-scaling needed for DRE systems, (2) hyper visors and data-center networks that do not support real-time scheduling of resources, and (3) absence of efficient and predictable fault tolerant mechanisms with acceptable overhead and consistency. This paper describes ongoing and proposed doctoral research to address these challenges.","PeriodicalId":447700,"journal":{"name":"2012 IEEE 31st Symposium on Reliable Distributed Systems","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129827702","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}
Replicated storage systems allow their stored data objects to outlive the life of the nodes storing them through replication. In this paper, we focus on durability, and more specifically on the concept of an object's lifetime, i.e., the duration of time between the creation of an object and when it is permanently irretrievable from the system. We analyze two main replication strategies: reactive, in which replication occurs in response to failures, and proactive, in which replication occurs in anticipation of failures. Our work presents a quantitative analysis that compares reactive and proactive through analytical models and simulations, considering exponentially distributed failures and reactive repairs, and periodic proactive replications. We also present a derivation of the analytical formula for the variance of the lifetime in the reactive model. Our results indicate that a proactive strategy leads to multiple times higher storage requirements than a reactive strategy. In addition, reactive systems are only moderately bursty in terms of bandwidth consumption, with rare peaks of at most five times the bandwidth consumption in proactive systems (given input parameter values that are compatible with real systems). Finally, for both strategies, the standard deviation is very close to the expected lifetime, and consequently, the lifetimes close to being exponentially distributed.
{"title":"A Quantitative Comparison of Reactive and Proactive Replicated Storage Systems","authors":"Rossana Motta, J. Pasquale","doi":"10.1109/SRDS.2012.1","DOIUrl":"https://doi.org/10.1109/SRDS.2012.1","url":null,"abstract":"Replicated storage systems allow their stored data objects to outlive the life of the nodes storing them through replication. In this paper, we focus on durability, and more specifically on the concept of an object's lifetime, i.e., the duration of time between the creation of an object and when it is permanently irretrievable from the system. We analyze two main replication strategies: reactive, in which replication occurs in response to failures, and proactive, in which replication occurs in anticipation of failures. Our work presents a quantitative analysis that compares reactive and proactive through analytical models and simulations, considering exponentially distributed failures and reactive repairs, and periodic proactive replications. We also present a derivation of the analytical formula for the variance of the lifetime in the reactive model. Our results indicate that a proactive strategy leads to multiple times higher storage requirements than a reactive strategy. In addition, reactive systems are only moderately bursty in terms of bandwidth consumption, with rare peaks of at most five times the bandwidth consumption in proactive systems (given input parameter values that are compatible with real systems). Finally, for both strategies, the standard deviation is very close to the expected lifetime, and consequently, the lifetimes close to being exponentially distributed.","PeriodicalId":447700,"journal":{"name":"2012 IEEE 31st Symposium on Reliable Distributed Systems","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128259819","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}
MapReduce scheme has became the state of the art in parallel processing of vast amount of data in distributed systems. Hadoop, as a popular open-source implementation of this technique, makes use of data block replication mechanism to provide a reliable and fault-tolerant design. To maintain data availability, Hadoop takes into account the possibilities of node and rack failures. Hence, it stores multiple copies of each data block to ensure availability and reliability. The current data block placement policy is to randomly distribute the replicas on all servers, satisfying some constraints such as preventing storage of two replicas of a data block on a single node. Our study proposes an efficient placement policy for data block replicas, which can reduce the consumed energy in data centers. The proposed policy is built upon the covering subset (CovSet) method. The effectiveness of the proposed approach is confirmed through simulations. Also, our experiments show that the proposed method becomes more effective whenever the average number of data blocks per server increases, which corresponds to the actual conditions in practice.
{"title":"Energy Efficient Hadoop Using Mirrored Data Block Replication Policy","authors":"Sara Arbab Yazd, S. Venkatesan, N. Mittal","doi":"10.1109/SRDS.2012.25","DOIUrl":"https://doi.org/10.1109/SRDS.2012.25","url":null,"abstract":"MapReduce scheme has became the state of the art in parallel processing of vast amount of data in distributed systems. Hadoop, as a popular open-source implementation of this technique, makes use of data block replication mechanism to provide a reliable and fault-tolerant design. To maintain data availability, Hadoop takes into account the possibilities of node and rack failures. Hence, it stores multiple copies of each data block to ensure availability and reliability. The current data block placement policy is to randomly distribute the replicas on all servers, satisfying some constraints such as preventing storage of two replicas of a data block on a single node. Our study proposes an efficient placement policy for data block replicas, which can reduce the consumed energy in data centers. The proposed policy is built upon the covering subset (CovSet) method. The effectiveness of the proposed approach is confirmed through simulations. Also, our experiments show that the proposed method becomes more effective whenever the average number of data blocks per server increases, which corresponds to the actual conditions in practice.","PeriodicalId":447700,"journal":{"name":"2012 IEEE 31st Symposium on Reliable Distributed Systems","volume":"295 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123389671","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}
Kossi Tiassou, K. Kanoun, M. Kaâniche, C. Seguin, Chris Papadopoulos
This paper addresses an aircraft mission operational reliability as resulting from component failures, environment changes, and maintenance facilities offered at the various stops involved in the mission. We will show how the on-line assessment of operational reliability will help adjust an aircraft mission, in case of major changes to equipment availability during the mission. The assessment is made possible thanks to the building and validation of a generic dependability model that is easily i) processed for the assignment of an initial mission, and ii) updated during mission accomplishment, following the occurrence of some specific major events. The generic model can be built as early as the design phase, by engineers who are specialist in dependability assessment, based on stochastic processes. Model update and processing, during aircraft operation, can be achieved by operators who are not necessarily familiar with stochastic processes in the way that they are being applied in this research. We will present examples of results that show the valuable role of operational dependability re-assessment during aircraft mission.
{"title":"Impact of Operational Reliability Re-assessment during Aircraft Missions","authors":"Kossi Tiassou, K. Kanoun, M. Kaâniche, C. Seguin, Chris Papadopoulos","doi":"10.1109/SRDS.2012.37","DOIUrl":"https://doi.org/10.1109/SRDS.2012.37","url":null,"abstract":"This paper addresses an aircraft mission operational reliability as resulting from component failures, environment changes, and maintenance facilities offered at the various stops involved in the mission. We will show how the on-line assessment of operational reliability will help adjust an aircraft mission, in case of major changes to equipment availability during the mission. The assessment is made possible thanks to the building and validation of a generic dependability model that is easily i) processed for the assignment of an initial mission, and ii) updated during mission accomplishment, following the occurrence of some specific major events. The generic model can be built as early as the design phase, by engineers who are specialist in dependability assessment, based on stochastic processes. Model update and processing, during aircraft operation, can be achieved by operators who are not necessarily familiar with stochastic processes in the way that they are being applied in this research. We will present examples of results that show the valuable role of operational dependability re-assessment during aircraft mission.","PeriodicalId":447700,"journal":{"name":"2012 IEEE 31st Symposium on Reliable Distributed Systems","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132601557","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}
Energy conservation and reliability of wireless communications are two crucial requirements of practical sensor networks. Radio duty cycling is a widely used mechanism to reduce energy consumption of sensor devices and to increase the lifetime of the network. A side effect of radio duty cycling is that it can cause the wireless communications to be unreliable---if a sender node transmits a packet while the receiver is asleep, the communication fails. Early duty cycling protocols like B-MAC that were designed for bit streaming radios achieve low duty cycle by keeping the radio transceiver awake for short time periods. However, they require a transmitter node to precede a packet transmission with a long preamble to ensure the reliability of wireless communication. Furthermore, they cannot be used with modern packet radios like widely used IEEE 802.15.4 based radio transceivers, which cannot transmit arbitrarily long preambles. Recent duty cycling schemes like X-MAC, on the other hand, reduce the length of the preamble and are designed to work with packet radios. However, in order to ensure that a receiver can reliably detect a transmitter's preamble transmission, these schemes need to turn the radio transceiver on for longer time durations than the early schemes like B-MAC. In this paper, we present a novel duty cycling scheme called Quick MAC, that achieves a very low duty cycle without compromising the reliability of wireless communication. Furthermore, Quick MAC is stateless, compatible with packet (and bit stream) radios, and does not require synchronization among sensor nodes. From our experiments using TMote sky motes, we show that Quick MAC reduces duty cycle by a factor of about 4 compared to X-MAC, and yet maintains the same level of reliability of wireless communication as X-MAC.
{"title":"Efficient Asynchronous Low Power Listening for Wireless Sensor Networks","authors":"R. Panta, James A. Pelletier, Gregg T. Vesonder","doi":"10.1109/SRDS.2012.23","DOIUrl":"https://doi.org/10.1109/SRDS.2012.23","url":null,"abstract":"Energy conservation and reliability of wireless communications are two crucial requirements of practical sensor networks. Radio duty cycling is a widely used mechanism to reduce energy consumption of sensor devices and to increase the lifetime of the network. A side effect of radio duty cycling is that it can cause the wireless communications to be unreliable---if a sender node transmits a packet while the receiver is asleep, the communication fails. Early duty cycling protocols like B-MAC that were designed for bit streaming radios achieve low duty cycle by keeping the radio transceiver awake for short time periods. However, they require a transmitter node to precede a packet transmission with a long preamble to ensure the reliability of wireless communication. Furthermore, they cannot be used with modern packet radios like widely used IEEE 802.15.4 based radio transceivers, which cannot transmit arbitrarily long preambles. Recent duty cycling schemes like X-MAC, on the other hand, reduce the length of the preamble and are designed to work with packet radios. However, in order to ensure that a receiver can reliably detect a transmitter's preamble transmission, these schemes need to turn the radio transceiver on for longer time durations than the early schemes like B-MAC. In this paper, we present a novel duty cycling scheme called Quick MAC, that achieves a very low duty cycle without compromising the reliability of wireless communication. Furthermore, Quick MAC is stateless, compatible with packet (and bit stream) radios, and does not require synchronization among sensor nodes. From our experiments using TMote sky motes, we show that Quick MAC reduces duty cycle by a factor of about 4 compared to X-MAC, and yet maintains the same level of reliability of wireless communication as X-MAC.","PeriodicalId":447700,"journal":{"name":"2012 IEEE 31st Symposium on Reliable Distributed Systems","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133232819","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}
In the recent years we have witnessed a proliferation of distributed stream processing systems that need to operate under bursty workloads. Examples include road traffic control, processing of financial feeds, network monitoring and real-time sensor data analysis systems. Meeting the QoS requirements of the stream processing systems under burstiness is a challenging process. In this paper we present our approach for adaptive rate allocation within the distributed stream processing system to meet the end-to-end execution time and rate demands of the applications. Our algorithm determines the rates of the application components at runtime, with respect to the QoS constraints, to compensate for delays experienced by the components or to react to sudden bursts of load. Our technique is distributed and low-cost. Our detailed experimental results over our Synergy middleware illustrate that our approach is practical, depicts good performance and has low resource overhead.
{"title":"RADAR: Adaptive Rate Allocation in Distributed Stream Processing Systems under Bursty Workloads","authors":"Ioannis Boutsis, V. Kalogeraki","doi":"10.1109/SRDS.2012.55","DOIUrl":"https://doi.org/10.1109/SRDS.2012.55","url":null,"abstract":"In the recent years we have witnessed a proliferation of distributed stream processing systems that need to operate under bursty workloads. Examples include road traffic control, processing of financial feeds, network monitoring and real-time sensor data analysis systems. Meeting the QoS requirements of the stream processing systems under burstiness is a challenging process. In this paper we present our approach for adaptive rate allocation within the distributed stream processing system to meet the end-to-end execution time and rate demands of the applications. Our algorithm determines the rates of the application components at runtime, with respect to the QoS constraints, to compensate for delays experienced by the components or to react to sudden bursts of load. Our technique is distributed and low-cost. Our detailed experimental results over our Synergy middleware illustrate that our approach is practical, depicts good performance and has low resource overhead.","PeriodicalId":447700,"journal":{"name":"2012 IEEE 31st Symposium on Reliable Distributed Systems","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114905019","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}
Ruijing Hu, Julien Sopena, L. Arantes, Pierre Sens, I. Demeure
We present a thorough performance comparison of three widely used probabilistic gossip algorithms over well-known random graphs. These graphs represent some large-scale network topologies: Bernoulli (or Erdos-Rényi) graph, random geometric graph, and scale-free graph. In order to conduct such a fair comparison, particularly in terms of reliability, we propose a new parameter, called effectual fan out. For a given topology and gossip algorithm, the effectual fan out characterizes the mean dissemination power of infected sites. For large-scale networks, the effectual fan out has thus a strong linear correlation with message complexity. It enables to make an accurate analysis of the behavior of a gossip algorithm over a topology. Furthermore, it simplifies the theoretical comparison of different gossip algorithms on the topology. Based on extensive experiments on top of OMNet++ simulator, which make use of the effectual fan out, we discuss the impact of topologies and gossip algorithms on performance, and how to combine them to have the best gain in terms of reliability.
{"title":"Fair Comparison of Gossip Algorithms over Large-Scale Random Topologies","authors":"Ruijing Hu, Julien Sopena, L. Arantes, Pierre Sens, I. Demeure","doi":"10.1109/SRDS.2012.28","DOIUrl":"https://doi.org/10.1109/SRDS.2012.28","url":null,"abstract":"We present a thorough performance comparison of three widely used probabilistic gossip algorithms over well-known random graphs. These graphs represent some large-scale network topologies: Bernoulli (or Erdos-Rényi) graph, random geometric graph, and scale-free graph. In order to conduct such a fair comparison, particularly in terms of reliability, we propose a new parameter, called effectual fan out. For a given topology and gossip algorithm, the effectual fan out characterizes the mean dissemination power of infected sites. For large-scale networks, the effectual fan out has thus a strong linear correlation with message complexity. It enables to make an accurate analysis of the behavior of a gossip algorithm over a topology. Furthermore, it simplifies the theoretical comparison of different gossip algorithms on the topology. Based on extensive experiments on top of OMNet++ simulator, which make use of the effectual fan out, we discuss the impact of topologies and gossip algorithms on performance, and how to combine them to have the best gain in terms of reliability.","PeriodicalId":447700,"journal":{"name":"2012 IEEE 31st Symposium on Reliable Distributed Systems","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116830895","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}
Anne-Marie Kermarrec, E. L. Merrer, G. Straub, Alexandre van Kempen
Distributed storage systems rely heavily on redundancy to ensure data availability as well as durability. In networked systems subject to intermittent node unavailability, the level of redundancy introduced in the system should be minimized and maintained upon failures. Repairs are well-known to be extremely bandwidth-consuming and it has been shown that, without care, they may significantly congest the system. In this paper, we propose an approach to redundancy management accounting for nodes heterogeneity with respect to availability. We show that by using the availability history of nodes, the performance of two important faces of distributed storage (replica placement and repair) can be significantly improved. Replica placement is achieved based on complementary nodes with respect to nodes availability, improving the overall data availability. Repairs can be scheduled thanks to an adaptive per-node timeout according to node availability, so as to decrease the number of repairs while reaching comparable availability. We propose practical heuristics for those two issues. We evaluate our approach through extensive simulations based on real and well-known availability traces. Results clearly show the benefits of our approach with regards to the critical trade-off between data availability, load-balancing and bandwidth consumption.
{"title":"Availability-Based Methods for Distributed Storage Systems","authors":"Anne-Marie Kermarrec, E. L. Merrer, G. Straub, Alexandre van Kempen","doi":"10.1109/SRDS.2012.10","DOIUrl":"https://doi.org/10.1109/SRDS.2012.10","url":null,"abstract":"Distributed storage systems rely heavily on redundancy to ensure data availability as well as durability. In networked systems subject to intermittent node unavailability, the level of redundancy introduced in the system should be minimized and maintained upon failures. Repairs are well-known to be extremely bandwidth-consuming and it has been shown that, without care, they may significantly congest the system. In this paper, we propose an approach to redundancy management accounting for nodes heterogeneity with respect to availability. We show that by using the availability history of nodes, the performance of two important faces of distributed storage (replica placement and repair) can be significantly improved. Replica placement is achieved based on complementary nodes with respect to nodes availability, improving the overall data availability. Repairs can be scheduled thanks to an adaptive per-node timeout according to node availability, so as to decrease the number of repairs while reaching comparable availability. We propose practical heuristics for those two issues. We evaluate our approach through extensive simulations based on real and well-known availability traces. Results clearly show the benefits of our approach with regards to the critical trade-off between data availability, load-balancing and bandwidth consumption.","PeriodicalId":447700,"journal":{"name":"2012 IEEE 31st Symposium on Reliable Distributed Systems","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130468739","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：