This document serves as the editorial introduction and table of contents for Volume 12, Issue 13 of the Proceedings of the VLDB Endowment (PVLDB), published in September 2019. This final issue of the volume features research destined for the 46th International Conference on Very Large Data Bases (VLDB 2020).
The PVLDB "Journal-Conference" Hybrid Model
The PVLDB model addresses the traditional "slow turnaround" problem of academic journals. By offering monthly submission deadlines and a rapid six-week review cycle, it provides the agility of online publication while maintaining the prestige and rigor of a top-tier conference. This issue highlights the success of this model in fostering a global data management community through high-quality peer review.
Research Frontiers: From Distributed Consistency to Trillion-Edge Graphs
The seven papers curated in this issue address the "scalability wall" across several critical domains:
- Distributed Systems: Research in this issue challenges the common industry trade-off where consistency is sacrificed for performance. One featured paper argues that "strong consistency is not hard to get," utilizing Two-Phase Locking and Two-Phase Commit protocols optimized for thousands of cores.
- Graph Analytics: There is a significant shift from processing millions of edges to "trillion-edge" graphs. The issue presents new techniques for distributed edge partitioning and online density bursting subgraph detection to handle this massive scale.
- Interactive Data Exploration: The move toward "progressive" data structures is a key insight. Instead of building indexes upfront, these structures evolve and improve as a user interacts with the data, enabling real-time interactivity.
- Data Quality and Storage: Other contributions include algorithmic discovery of uniqueness constraints for fact-checking and optimal column layouts for hybrid transactional and analytical workloads (HTAP).
Editorial Leadership and Peer Review
The success of PVLDB relies on a robust editorial structure. The process involves a large team of expert reviewers and associate editors to ensure high-quality feedback and rapid dissemination of research.
The issue was overseen by a dedicated team, including Publication Editors Abdul Quamar and Yongxin Tong, and Managing Editor Wolfgang Lehner, supported by an Advisory Committee of international experts from institutions such as Amazon, CNRS, and the University of Toronto.
Conclusion
This final issue of Volume 12 serves as a catalyst for future work in advanced analytics and distributed data management. The editors hope these seven papers inspire novel systems and algorithmic contributions that continue to push the boundaries of what is possible with very large data bases.
Summary updated