ResLake: Towards Minimum Job Latency and Balanced Resource Utilization in Geo-distributed Job Scheduling
Summary: ResLake is a global geo-distributed scheduler that jointly minimizes job completion time while balancing multi-resource (CPU/memory/disk) utilization across data centers, explicitly trading off WAN transfer cost against latency instead of minimizing WAN alone. Deployed at ByteDance for 1.5 years managing billions of jobs, ResLake cuts JCT by ≥20% and improves cross‑DC resource balance up to 53%. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Xinchun Zhang
- 2. Aqsa Kashaf
- 3. Yihan Zou
- 4. Wei Zhang
- 5. Weibo Liao
- 6. Haoxiang Song
- 7. Jintao Ye
- 8. Yakun Li
- 9. Rui Shi
- 10. Yong Tian
- 11. Wei Feng
- 12. Binbin Chen
- 13. Zuzhi Chen
- 14. Tieying Zhang
- 15. Yongping Tang
Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 3 of 3 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 209 | Schism: a Workload-Driven Approach to Database Replication and Partitioning | 2010 | VLDB | 0.00034468292 |
| 4,046 | WANalytics: Geo-Distributed Analytics for a Data Intensive World | 2015 | SIGMOD | 6.4979392e-05 |
| 5,185 | Yugong: Geo-Distributed Data and Job Placement at Scale | 2019 | VLDB | 5.6405374e-05 |
Previous
Page 1 / 1
Next