{"title":"Shared Data from a Study of Measurement Uncertainty in Fault Injection","authors":"R. Barbosa, D. Skarin, J. Karlsson","doi":"10.1109/ICDCSW.2010.68","DOIUrl":"https://doi.org/10.1109/ICDCSW.2010.68","url":null,"abstract":"Experimental dependability studies usually produce an amount of data substantially greater than what can be presented in a research paper or a technical report. For this reason, authors condensate the results into more succinct forms that allow them to convey their message. Since a large amount of the original data is left unexplored, sharing it allows other teams to discover additional facts (as well as to compare the results to other studies). In a previous paper, we investigated sources of uncertainty in measurement results obtained using three different fault injection techniques. The resulting experimental data was shared in the AMBER raw data repository. This paper gives an overview of the study and makes an attempt at further exploring the shared data.","PeriodicalId":133907,"journal":{"name":"2010 IEEE 30th International Conference on Distributed Computing Systems Workshops","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129924907","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}
{"title":"Towards an Efficient QoS Based Selection of Neighbors in QOLSR","authors":"F. Khadar, N. Mitton, D. Simplot-Ryl","doi":"10.1109/ICDCSW.2010.50","DOIUrl":"https://doi.org/10.1109/ICDCSW.2010.50","url":null,"abstract":"OLSR is a well-known proactive protocol for wireless networks. Although very efficient by many points, it suffers from the drawbacks of not taking into account QoS metrics such as delay or bandwidth. To overcome this pitfall, some QOLSR (QoS OLSR) solutions have been designed. Nevertheless, they still provide weak performance regarding QoS metrics. In this paper, we introduce a novel and simple neighbor selection to allow routing with OLSR along quasi-optimal QoSpaths. Simulation results show that our solution outperforms solutions from the literature by providing efficient QoS paths (less than $2%$overhead from optimal paths) by selecting a very small set of neighbors to route packets.","PeriodicalId":133907,"journal":{"name":"2010 IEEE 30th International Conference on Distributed Computing Systems Workshops","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130010897","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}
M. Paone, A. Cucinotta, Antonino Longo Minnolo, L. Paladina, A. Puliafito, A. Zaia
{"title":"A Bio-inspired Distributed Routing Protocol for Wireless Sensor Networks: Performance Evaluation","authors":"M. Paone, A. Cucinotta, Antonino Longo Minnolo, L. Paladina, A. Puliafito, A. Zaia","doi":"10.1109/ICDCSW.2010.66","DOIUrl":"https://doi.org/10.1109/ICDCSW.2010.66","url":null,"abstract":"Sensor networks are wireless networks with peculiar properties that require specific attention to be properly managed if their performance and dependability have to be taken care of. In fact, both the usually hard environmental conditions where sensors are deployed and the scarcity of node resources determine overall critical working conditions that dramatically impact on parameters such as reliability, fault tolerance, power consumption and scalability. Currently available routing protocols hardly meet all these constraints, because they are not designed in a strong distributed fashion. Swarm intelligence is a science that studies the behavior of social insects characterized by distributed elaboration features, simple computational capacity, flexibility and robustness. In this paper, we propose a routing algorithm based on swarm intelligence, named pheromone protocol (PH-protocol), with some interesting features such as self organization, fault tolerance and environmental adaptation capabilities. To highlight these properties we compared our algorithm against the directed diffusion one, a well known sensor network routing protocol, and measured their performances in terms of reliability, fault tolerance, power consumption and scalability. The obtained results demonstrate that sensor networks greatly benefit from the distributed nature of our PH-protocol, which ensures interesting performance and reliability features.","PeriodicalId":133907,"journal":{"name":"2010 IEEE 30th International Conference on Distributed Computing Systems Workshops","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124447633","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. D. Stefani, P. Gamba, E. Goldoni, A. Savioli, Davide Silvestri, Flavio Toffalini
{"title":"REnvDB, a RESTful Database for Pervasive Environmental Wireless Sensor Networks","authors":"F. D. Stefani, P. Gamba, E. Goldoni, A. Savioli, Davide Silvestri, Flavio Toffalini","doi":"10.1109/ICDCSW.2010.56","DOIUrl":"https://doi.org/10.1109/ICDCSW.2010.56","url":null,"abstract":"Every application of a Wireless Sensor Network (WSN) requires the data collected from sensor nodes to be reachable from the outside, in order to allow its processing and to obtain information from the monitored phenomena. Environmental protection and agriculture are examples of areas most likely to benefit from the deployment of WSNs. Devices equipped with temperature, humidity and light sensors could play a role in preserving the environment by providing early alerts and useful status information in real-time. Since such networks usually provide a huge amount of raw data, an easy way to organize, store and retrieve useful information is also needed. In this work we present REnvDB (RESTful Environmental DataBase), a novel database for Wireless Sensor Networks specifically designed for pervasive monitoring applications. The data model of the database is flexible enough to fulfill the requirements of most of the typical environmental monitoring systems. We also provided the database a RESTful interface, which can be used to expose the data collected by heterogeneous WSNs using the standard protocols of the World Wide Web. Thanks to its modular design and the use of metadata, the behavior of our RESTful interface can also be easily extended or modified according to the requirements and constraints of a specific context. Finally, the entire system has been implemented and tested in a real environment as part of a bigger project for the monitoring of an intensive agricultural field.","PeriodicalId":133907,"journal":{"name":"2010 IEEE 30th International Conference on Distributed Computing Systems Workshops","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121451744","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}
{"title":"Physical Layer Network Isolation in Multi-tenant Clouds","authors":"Jack Brassil","doi":"10.1109/ICDCSW.2010.39","DOIUrl":"https://doi.org/10.1109/ICDCSW.2010.39","url":null,"abstract":"The isolation capabilities provided by conventional enterprise data center technology are inadequate for many clients of multi-tenant storage or compute clouds. To address this deficiency we propose a cloud architecture which relies on strategic deployment of physical layer network isolation. We show how improved client isolation can be realized with 'color gapping' using existing Coarse Wavelength Division Multiplexing technologies, and demonstrate how in some settings isolation can be visually verified by clients. To address the challenging problem of isolating multiple clients' Virtual Machines within a single physical server, we introduce a novel application of distributed bridging based on Virtual Ethernet Port Aggregators.","PeriodicalId":133907,"journal":{"name":"2010 IEEE 30th International Conference on Distributed Computing Systems Workshops","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115767837","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}
{"title":"Field Data Based Modeling of Sender Based Message Logging Protocols for Supercomputers Checkpointing","authors":"Domenico Cotroneo, C. Martino","doi":"10.1109/ICDCSW.2010.79","DOIUrl":"https://doi.org/10.1109/ICDCSW.2010.79","url":null,"abstract":"Checkpointing is today’s common mean for dealing with transient failures in supercomputers. However, the effectiveness of checkpointing and recovery protocols under the assumption that failures may happen during their operation is not well understood. We present an evaluation of the checkpointing and recovery based on Sender Based Message Logging protocols (SBML). We evaluate it by means of a model which is gathered from an extensive field data campaign performed on the SCOPE supercomputer at the University of Naples. A comprehensive model is built to evaluate reliability, scalability and performance of SBML. The proposed model takes into account failures during the checkpointing and recovery. Result provide insights on the limit of the number of nodes that can be allocated to the same job considering i) the overhead for the distributed coordination for rollback, and ii) network resources (e.g. bandwidth).","PeriodicalId":133907,"journal":{"name":"2010 IEEE 30th International Conference on Distributed Computing Systems Workshops","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122064760","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}
{"title":"From Performance to Resilience Benchmarking","authors":"Raquel Almeida, H. Madeira, M. Vieira","doi":"10.1109/ICDCSW.2010.69","DOIUrl":"https://doi.org/10.1109/ICDCSW.2010.69","url":null,"abstract":"In this paper we discuss the resilience benchmarking concept. First we present a brief historical overview on benchmarking. Then we discuss the key components of a resilience benchmarking. Finally, we propose a research path for tackling the resilience benchmarking challenge.","PeriodicalId":133907,"journal":{"name":"2010 IEEE 30th International Conference on Distributed Computing Systems Workshops","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123615354","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}
{"title":"Mobile Software Agents for Wireless Network Mapping and Dynamic Routing","authors":"Hamzeh Khazaei, J. Misic, V. Mišić","doi":"10.1109/ICDCSW.2010.65","DOIUrl":"https://doi.org/10.1109/ICDCSW.2010.65","url":null,"abstract":"Mapping and dynamic routing in a wireless network are essential and basic operations that are the prerequisites for all higher order interactions between nodes on that network. The Intent of network mapping by multi mobile software agents is to obtain the topology of network in a distributed manner. In this paper we examine how software agents can wander in an unknown Ad hoc network with cooperation to report the topology of the network. We present results from a simulation of such a system and discuss the relationship between the overall efficiency of the system and diversity of the agent types, agent population, agent memory, inter-agent communication styles. In dynamic network routing, mobile agents are responsible for updating the routing table of nodes for a multi-hop communication purpose. The dynamic nature of the topology of the Ad hoc networks is due to mobility of some nodes in the network: wireless links are broken and reformed frequently. In this paper we present a dynamic, wireless, peer to peer network with routing tasks performed in a decentralized and distributed fashion by mobile software agents that cooperate to accumulate and distribute network connectivity information. We also examine different types of agents, stigmergic versus non stigmergic having identical overheads to study the outcome connectivity in a dynamic wireless network.","PeriodicalId":133907,"journal":{"name":"2010 IEEE 30th International Conference on Distributed Computing Systems Workshops","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125157068","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}
{"title":"A Few Bytes are Worth a Thousand Words: Run-Time Compilation of High Level Scripts in Sensor Networks","authors":"J. Ellul, K. Martinez","doi":"10.1109/ICDCSW.2010.51","DOIUrl":"https://doi.org/10.1109/ICDCSW.2010.51","url":null,"abstract":"Over the air reprogramming has become a vital requirement of wireless sensor networks due to the infeasibility and inaccessibility of deployed sensor nodes. The transmission of whole software binaries consumes a substantial amount of energy and therefore a more efficient means of updating software on sensor nodes is required. Virtual machines allow for smaller sized update scripts due to the higher level instructions. However, they also incur higher execution overheads introduced by the required interpretation. Native code on the other hand provides faster execution however at the cost of a larger code size. In this paper we demonstrate that it is possible to leverage small update sizes achieved using high level scripts without incurring high interpretation overheads, by translating the scripts to native code in situ.","PeriodicalId":133907,"journal":{"name":"2010 IEEE 30th International Conference on Distributed Computing Systems Workshops","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131496038","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}
{"title":"Multiple Micro Virtual Disks Based on NaryRAID for Large Scale Storage","authors":"Yuji Nakamura, M. Uehara","doi":"10.1109/ICDCSW.2010.48","DOIUrl":"https://doi.org/10.1109/ICDCSW.2010.48","url":null,"abstract":"If we do not use only idle resources of servers but also those of clients, we can reduce the cost of private cloud. In the previous works, we proposed VLSD(Virtual Large Scale Disks) as a toolkit for constructing large scale storage such as private cloud. VLSD provides several RAID classes. Using VLSD, we developed complex software RAID named Single Large Virtual Disk (SLVD). However, since it has a centralized file server, it becomes the single point of failure and the bottle neck. In this paper, we propose a novel Multiple Micro Virtual Disks (MMVD) in order to avoid such issues. In addition, we apply NaryRAID, 3 fault tolerant RAID to MMVD. NaryRAID based MMVD achieves higher reliability than RAID6 based one.","PeriodicalId":133907,"journal":{"name":"2010 IEEE 30th International Conference on Distributed Computing Systems Workshops","volume":"6 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134582659","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}