The part-time parliament

Lamport, Leslie

doi:10.1145/3335772.3335939

Cited by 390 publications

(604 citation statements)

References 0 publications

Supporting

Mentioning

578

Contrasting

Unclassified

Order By: Relevance

“…These protocols allow any minority of sites to fail simultaneously: e.g., running a typical protocol over 13 data centers would tolerate 6 of them failing. However, natural disasters leading to the loss of a data center are rare, and planned downtime can be handled by reconfiguring the unavailable site out of the system [15,31]. Furthermore, temporary data center outages (e.g., due to connectivity issues) typically have a short duration [20], and, as we confirm experimentally in §5, rarely happen concurrently.…”

Section: Introductionmentioning

confidence: 92%

“…The classical way of implementing SMR is by funneling all commands through a single leader replica [14,15,26,27], which impairs scalability. A way to mitigate this problem is to distribute the leader responsibilities round-robin among replicas, as done in Mencius [21].…”

Section: Related Workmentioning

confidence: 99%

“…Unfortunately, existing SMR protocols are poorly suited to planet-scale systems. Common SMR protocols, such as Paxos [15] and Raft [27], are rooted in cluster computing where a leader site determines the ordering of commands. This is unfair to clients far away from the leader.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

State-machine replication for planet-scale systems

Enes

Baquero

Rezende

et al. 2020

Proceedings of the Fifteenth European Conference on Computer Systems

View full text Add to dashboard Cite

Online applications now routinely replicate their data at multiple sites around the world. In this paper we present Atlas, the first state-machine replication protocol tailored for such planet-scale systems. Atlas does not rely on a distinguished leader, so clients enjoy the same quality of service independently of their geographical locations. Furthermore, clientperceived latency improves as we add sites closer to clients. To achieve this, Atlas minimizes the size of its quorums using an observation that concurrent data center failures are rare. It also processes a high percentage of accesses in a single round trip, even when these conflict. We experimentally demonstrate that Atlas consistently outperforms state-of-the-art protocols in planet-scale scenarios. In particular, Atlas is up to two times faster than Flexible Paxos with identical failure assumptions, and more than doubles the performance of Egalitarian Paxos in the YCSB benchmark.

show abstract

Section: Introductionmentioning

confidence: 92%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

State-machine replication for planet-scale systems

Enes

Baquero

Rezende

et al. 2020

Proceedings of the Fifteenth European Conference on Computer Systems

View full text Add to dashboard Cite

show abstract

“…Paxos [Lamport 1998[Lamport , 2006] is a distributed consensus algorithm for asynchronous distributed systems. It can be used to create replicas of distributed services to ensure greater availability of the service as a whole [Schneider 1990].…”

Section: Introductionmentioning

confidence: 99%

“…However, accesses to stable memory are usually much slower than accesses to main memory, and algorithms in the fail-recovery model minimize the use of this costly resource. For example, Paxos has two writes to stable memory in its critical path [Lamport 1998]. This way, performance of distributed algorithms in the fail-recovery model will lean heavily on the performance of stable memory.…”

Section: Introductionmentioning

confidence: 99%

Characterizing Synchronous Writes in Stable Memory Devices

Mingardi

Vieira

2019

Anais Do Workshop Em Desempenho De Sistemas Computacionais E De Comunicação (WPerformance)

View full text Add to dashboard Cite

Distributed algorithms that operate in the fail-recovery model rely on the state stored in stable memory to guarantee the irreversibility of operations even in the presence of failures. The performance of these algorithms lean heavily on the performance of stable memory. Current storage technologies have a defined performance profile: data is accessed in blocks of hundreds or thousands of bytes, random access to these blocks is expensive and sequential access is somewhat better. File system implementations hide some of the performance limitations of the underlying storage devices using buffers and caches. However, fail-recovery distributed algorithms bypass some of these techniques and perform synchronous writes to be able to tolerate a failure during the write itself. Assuming the distributed system designer is able to buffer the algorithm's writes, we ask how buffer size and latency complement each other. In this paper we start to answer this question by characterizing the performance (throughput and latency) of typical stable memory devices using a representative set of current file systems.

show abstract

Staged data delivery protocol: A blockchain‐based two‐stage protocol for non‐repudiation data delivery

Wang

Liu

2021

Concurrency and Computation

View full text Add to dashboard Cite

For many years, non‐repudiation in data delivery has not been satisfactorily achieved due to the difficulty of supervising delivery behavior. Existing fair protocols are still not entirely free of third parties who are usually not as trustworthy as expected. Moreover, most of them are built on complex encryption schemes, resulting in inefficiencies when handling large chunks of data and non‐exchanged data sharing. To be able to deliver data without third parties and repudiation, we propose the staged data delivery protocol (SDDP) that integrates fair data delivery into Blockchain consensus processes. SDDP confirms transactions of data delivery in a cascading manner where nodes sign on each transaction one after another. It consists of two stages, in which neither party will have a considerable advantage over the other. In this way, SDDP enables non‐repudiation of directly delivering large chunks of data through an understandable communication strategy; and the Blockchain network replaces third parties to preserve an immutable history of data delivery activities. The results of extensive evaluations show that SDDP is admissible for dealing with fair data delivery on the basis of its good qualities and performance.

show abstract

The part-time parliament

Cited by 390 publications

References 0 publications

State-machine replication for planet-scale systems

State-machine replication for planet-scale systems

Characterizing Synchronous Writes in Stable Memory Devices

Staged data delivery protocol: A blockchain‐based two‐stage protocol for non‐repudiation data delivery

Contact Info

Product

Resources

About