Database Paper Browser

Back to papers

Towards Resource Efficiency: Practical Insights into Large-Scale Spark Workloads at ByteDance

Summary: Governance for large-scale Spark: push-based Cloud Shuffle + ESS to cut I/O stalls, and extended configs (milliCores, memoryBurst, spill modes) for fine-grained resource control. Two-stage auto-tuning on millions of jobs yields +22% CPU, +5% memory, 10% less shuffle time. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID
13581
Venue
VLDB
Year
2024
Pagerank
4.3849295e-05
Overall Rank
9,155 | 36.32%
DOI
10.14778/3685800.3685804

Incoming Non-self Citations Over Time

Authors

Incoming Citations (Sorted by Pagerank)

Showing 2 of 2 citing papers.

Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 7 of 7 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Previous Page 1 / 1 Next

Semantically Similar Papers