Fast Log Replication in Highly Available Data Store

Wang, Donghui; Cai, Peng; Qian, Weining; Zhou, Aoying; Pang, Tianze; Jiang, Jing

doi:10.1007/978-3-319-63564-4_20

Cited by 4 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our method for selecting, evaluating, filtering, and testing data items, and in turn building the data corpus, has substantial practical benefits as it enabled us to obtain a consistent data set in a relatively short time, with the methodological rigor required for qualitative research study. By utilizing big-data tools for building a thick data set (Latzko-Toth et al, 2017; Wang, 2016), we were able to reach both the breadth and scope of big-data collection as well as the holistic, meaningful, and context-grounded depth of small-scale qualitative analysis.…”

Section: Discussionmentioning

confidence: 99%

Whose dystopia is it anyway? Deepfakes and social media regulation

Yadlin-Segal

Oppenheim

2020

Convergence: The International Journal of Research into New Med

View full text Add to dashboard Cite

This study explores global journalistic discussions of deepfake applications (audiovisual manipulating applications based on artificial intelligence (AI)) to understand the narratives constructed through global coverage, the regulatory actions associated with these offered narratives, and the functions such narratives might serve in global sociopolitical contexts. Through a qualitative–interpretive narrative analysis, this article shows how journalists frame deepfakes as a destabilizing platform that undermines a shared sense of social and political reality, enables the abuse and harassment of women online, and blurs the acceptable dichotomy between real and fake. This phenomenon is tied to discussions of dis/misinformation, manipulation, exploitation, and polarization in the media ecosystem these days. Based on these findings, the article then provides broader practical and theoretical insights about AI content regulation and ethics, accountability, and responsibility in digital culture.

show abstract

Section: Discussionmentioning

confidence: 99%

Whose dystopia is it anyway? Deepfakes and social media regulation

Yadlin-Segal

Oppenheim

2020

Convergence: The International Journal of Research into New Med

View full text Add to dashboard Cite

show abstract

“…As mentioned in Section II.A, when the new log is applied in the leader, using the pipeline to broadcast it to the followers is a widely used implementation method of the SMR protocol [29]- [31]. Therefore, researchers designed the log buffer to merges the logs and replicate them to each follower node, which can significantly reduce the RPC connections between nodes, thereby achieving the effect of improving the TPS [32], [33].…”

Section: B Related Workmentioning

confidence: 99%

Leader Confirmation Replication for Millisecond Consensus in Private Chains

Zhu,

Du,

Sun

et al. 2021

Preprint

View full text Add to dashboard Cite

Geo-distributed private chain and database have created higher performance requirements for consistency models. However, with millisecond network latency between nodes, the widely used leader-based SMR models cause frequent retransmission of logs since they cannot know the logs replication status in time, which resulting in the leader costing high network and computing resource. To address the problem, we proposed a Leader Confirmation based Replication (LCR) model. First, we demonstrate the efficacy of the approach by designing the Future-Log Replication model, which the followers are responsible for non-transactional log replication. It reduces the leader's network load using the signal log. Secondly, we designed a Generation Rereplication strategy, which can ensure the security and consistency of future-logs when the number of nodes changes. Finally, we implemented LCR-Raft and designed experiments. The results show that in the single-ms network latency environments, LCR-Raft can provide higher TPS (1.5X 1.9X), and reduce the network traffic of the leader by 20%-30% with acceptable network traffic and CPU cost on followers. Besides, LCR can provide high portability since it does not change the number of leader and election process.

show abstract