Filters








179 Hits in 4.6 sec

Extending Classic Paxos for High-performance Read-Modify-Write Registers [article]

Vasilis Gavrielatos, Antonios Katsarakis, Vijay Nagarajan
2021 arXiv   pre-print
In this work we provide a detailed specification of how we extended and implemented Classic Paxos (CP) to execute Read-Modify-Writes.  ...  In addition, we also specify how we implemented All-aboard Paxos over CP and how we use carstamps, to also add ABD reads and writes, to accelerate the common case, where RMWs are not needed.  ...  In this work we provide a detailed specification of how we extended and implemented Classic Paxos to execute Read-Modify-Writes in Kite [2] .  ... 
arXiv:2103.14701v1 fatcat:u3y2ylr43fcnrghloabyne2pzy

The Impact of RDMA on Agreement

Marcos K. Aguilera, Naama Ben-David, Rachid Guerraoui, Virendra Marathe, Igor Zablotchi
2019 Proceedings of the 2019 ACM Symposium on Principles of Distributed Computing - PODC '19  
This technology allows a process to directly read and write the memory of a remote host, with a mechanism to control access permissions.  ...  Specifically, we show that RDMA allows algorithms that simultaneously achieve high resilience and high performance, while traditional algorithms had to choose one or another.  ...  We wish to thank the anonymous reviewers for their helpful comments on improving the paper.  ... 
doi:10.1145/3293611.3331601 dblp:conf/podc/AguileraBGMZ19 fatcat:ijm6ini3lrahpgzvorlodnk6e4

Evaluating Byzantine Quorum Systems

Wagner Saback Dantas, Alysson Neves Bessani, Joni da Silva Fraga, Miguel Correia
2007 2007 26th IEEE International Symposium on Reliable Distributed Systems (SRDS 2007)  
For that purpose, a framework for evaluating BQS called BQSNeko was developed. The results of the evaluation allow a better understanding of the algorithms and the tradeoffs involved.  ...  Many BQS have been proposed recently, but comparing their performance is not simple.  ...  We thank Eduardo Alchieri for his contribution on the Byzantine Paxos implementation.  ... 
doi:10.1109/srds.2007.34 dblp:conf/srds/DantasBFC07 fatcat:r6yuguijqne35ihamb57ulyhia

Evaluating Byzantine Quorum Systems

Wagner Saback Dantas, Alysson Neves Bessani, Joni da Silva Fraga, Miguel Correia
2007 Symposium on Reliable Distributed Systems. Proceedings  
For that purpose, a framework for evaluating BQS called BQSNeko was developed. The results of the evaluation allow a better understanding of the algorithms and the tradeoffs involved.  ...  Many BQS have been proposed recently, but comparing their performance is not simple.  ...  We thank Eduardo Alchieri for his contribution on the Byzantine Paxos implementation.  ... 
doi:10.1109/srds.2007.4365701 fatcat:zffrdmkn5ffa3nvmhyarsp7hmu

Paxos Consensus, Deconstructed and Abstracted [chapter]

Álvaro García-Pérez, Alexey Gotsman, Yuri Meshman, Ilya Sergey
2018 Lecture Notes in Computer Science  
Lamport's Paxos algorithm is a classic consensus protocol for state machine replication in environments that admit crash failures.  ...  Many versions of Paxos exploit the protocol's intrinsic properties for the sake of gaining better run-time performance, thus widening the gap between the original description of the algorithm, which was  ...  We thank the ESOP 2018 reviewers for their feedback. We also thank Amal Ahmed for her efforts as ESOP 2018 Programme Chair.  ... 
doi:10.1007/978-3-319-89884-1_32 fatcat:t5cqoiq6czayjps2r6eute4un4

GIRAFFE: A scalable distributed coordination service for large-scale systems

Xuanhua Shi, Haohong Lin, Hai Jin, Bing Bing Zhou, Zuoning Yin, Sheng Di, Song Wu
2014 2014 IEEE International Conference on Cluster Computing (CLUSTER)  
We evaluate Giraffe on a high performance computing test-bed. The experimental results show that Giraffe gains much better write performance than ZooKeeper when server ensemble is large.  ...  There are three important contributions in our design. (1) Giraffe organizes coordination servers using interior-nodedisjoint trees for better scalability. (2) Giraffe employs a novel Paxos protocol for  ...  ACKNOWLEDGMENTS We would like to thank the anonymous reviewers, Franck Cappello, Qi Huang, Thomas Ropars, Chengwei Wang, and Xiaobo Zhou for their valuable Comments, as well as Iu-  ... 
doi:10.1109/cluster.2014.6968766 dblp:conf/cluster/ShiLJZYDW14 fatcat:3ftxgk6nanaabdt3bbwrfhbmki

MDCC: Multi-Data Center Consistency [article]

Tim Kraska, Gene Pang, Michael J. Franklin, Samuel Madden
2012 arXiv   pre-print
Replicating data across multiple data centers not only allows moving the data closer to the user and, thus, reduces latency for applications, but also increases the availability in the event of a data  ...  Synchronous wide-area replication is therefore considered to be unfeasible with strong consistency and current solutions either settle for asynchronous replication which implies the risk of losing data  ...  If they are not equal, the value was modified between the read and write and a write-write conflict was encountered.  ... 
arXiv:1203.6049v1 fatcat:vorf7zi67jgnbgz4rklqupn6vu

Paxos Consensus, Deconstructed and Abstracted (Extended Version) [article]

Álvaro García-Pérez, Alexey Gotsman, Yuri Meshman, Ilya Sergey
2018 arXiv   pre-print
Lamport's Paxos algorithm is a classic consensus protocol for state machine replication in environments that admit crash failures.  ...  Many versions of Paxos exploit the protocol's intrinsic properties for the sake of gaining better run-time performance, thus widening the gap between the original description of the algorithm, which was  ...  We thank the ESOP 2018 reviewers for their feedback. This work by was supported by ERC Starting Grant H2020-EU 714729 and EPSRC First Grant EP/P009271/1.  ... 
arXiv:1802.05969v1 fatcat:roxatjzdcfczhim6elusi26aoi

Disk Paxos

Eli Gafni, Leslie Lamport
2003 Distributed computing  
We present an algorithm, called Disk Paxos, for implementing a reliable distributed system with a network of processors and disks.  ...  Like the original Paxos algorithm, Disk Paxos maintains consistency in the presence of arbitrary non-Byzantine faults.  ...  register; and it reads the values of dblock [q] for other processors q from the registers.  ... 
doi:10.1007/s00446-002-0070-8 fatcat:fumtprnclnfc5fqpfb3gcdsiiu

Disk Paxos [chapter]

Eli Gafni, Leslie Lamport
2000 Lecture Notes in Computer Science  
We present an algorithm, called Disk Paxos, for implementing a reliable distributed system with a network of processors and disks.  ...  Like the original Paxos algorithm, Disk Paxos maintains consistency in the presence of arbitrary non-Byzantine faults.  ...  register; and it reads the values of dblock [q] for other processors q from the registers.  ... 
doi:10.1007/3-540-40026-5_22 fatcat:7fgfkftqlvholfjaxyt3ms3cj4

NetChain: Scale-Free Sub-RTT Coordination (Extended Version) [article]

Xin Jin, Xiaozhou Li, Haoyu Zhang, Nate Foster, Jeongkeun Lee, Robert Soule, Changhoon Kim, Ion Stoica
2018 arXiv   pre-print
The major challenge is to achieve low latency and high throughput while providing strong consistency and fault-tolerance.  ...  Acknowledgments We thank our shepherd Amar Phanishayee and the anonymous reviewers for their valuable feedback. Robert Soulé is supported in part by SNF 167173.  ...  Modern programmable switch ASICs (e.g., Barefoot Tofino [13] ) provide onchip register arrays to store user-defined data that can be read and modified for each packet at line rate.  ... 
arXiv:1802.08236v1 fatcat:3lav3tnuejbqlhrsj4hjraxpkm

Photon

Rajagopal Ananthanarayanan, Shivakumar Venkataraman, Venkatesh Basker, Sumit Das, Ashish Gupta, Haifeng Jiang, Tianhao Qiu, Alexey Reznichenko, Deomid Ryabkov, Manpreet Singh
2013 Proceedings of the 2013 international conference on Management of data - SIGMOD '13  
In this paper, we describe the architecture of Photon, a geographically distributed system for joining multiple continuously flowing streams of data in real-time with high scalability and low latency,  ...  It produces joined logs that are used to derive key business metrics, including billing for advertisers.  ...  The maximum number of simultaneously tolerated datacenter outages should be configurable. • Read-modify-write transactions: Workers must be able to perform conditional commits such as writing an event  ... 
doi:10.1145/2463676.2465272 dblp:conf/sigmod/AnanthanarayananBDGJQRRSV13 fatcat:mpn26iqvrve6plkc6qwndjhshm

Odyssey

Vasilis Gavrielatos, Antonios Katsarakis, Vijay Nagarajan
2021 Proceedings of the Sixteenth European Conference on Computer Systems  
In this paper, we investigate the impact of modern hardware on the performance of strongly-consistent replication protocols.  ...  First, we create an informal taxonomy of replication protocols, based on which we carefully select 10 protocols for analysis.  ...  Acknowledgments We would like to thank Boris Grot, the anonymous reviewers and our shepherd Vivien Quema for their valuable comments and feedback.  ... 
doi:10.1145/3447786.3456240 fatcat:wswdm5gavrcpdistrwrh4fy43e

Beernet

B. Mejías, P. Van Roy
2010 International Journal of Adaptive, Resilient and Autonomic Systems  
We present a decentralized system built with a structured overlay network which is self-organized and self-healing, providing a transactional replicated storage for small or large scale systems.  ...  All peers responsible for the items involved in Algorithm 1 Swap transaction BOT x = read(item i); y = read(item j); write(item j, x); write(item i, y) ; EOT the transaction, as well as their replicas,  ...  We have adapted Paxos to support eager locking adding a notification mechanism for the registered readers of every shared item.  ... 
doi:10.4018/jaras.2010070101 fatcat:of5pt4kddbhxhddgunxv2zp2ly

HovercRaft

Marios Kogias, Edouard Bugnion
2020 Proceedings of the Fifteenth European Conference on Computer Systems  
Adding nodes to a system can therefore either increase performance at the expense of consistency, or increase resiliency at the expense of performance.  ...  , linear speedup over unreplicated configuration for selected workloads, and a 4× speedup for the YCSBE-E benchmark running on Redis over an unreplicated deployment.  ...  Acknowledgements We would like to thank the anonymous PC and shadow-PC reviewers, our shepherd Marco Canini, and Jim Larus, for their valuable feedback on the paper.  ... 
doi:10.1145/3342195.3387545 dblp:conf/eurosys/KogiasB20 fatcat:6ixq2egsyvaonisqvs5bgjnxcq
« Previous Showing results 1 — 15 out of 179 results