Skip to main content
Engineering, Data / ML, Uber AI

Capacity Recommendation Engine: Throughput and Utilization Based Predictive Scaling

19 January 2022 / Global
Featured image for Capacity Recommendation Engine: Throughput and Utilization Based Predictive Scaling
Figure 1: Golden Signal Metrics Utilized by CRE
Figure 2: Throughput Decomposition Result
Figure 3: Utilization vs Throughput per Core
Figure 4: Utilization vs Throughput per Core
Figure 5: Service Target TPC Trend
Figure 6: Scheduled Flow
Figure 7: On-demand Flow
Figure 8: Data Ingestion Flow
Figure 9: Region Capacity
Figure 10: Region Utilization
Shu-Ming Peng

Shu-Ming Peng

Shu-Ming Peng is a Sr. Software Engineer on Uber's Maps Production Engineering team based in Sunnyvale, CA. Shu-Ming works on building automated solutions to improve system reliability and efficiency.

Jianing He

Jianing He

Jianing He is a Software Engineer on Uber’s Maps Production Engineering team in San Francisco, CA. Jianing primarily works on improving Uber services reliability and efficiency with data-driven capacity and performance analysis tools.

Ranjib Dey

Ranjib Dey

Ranjib Dey is a Staff Software Engineer on Uber’s Maps Production Engineering team in San Francisco, CA. Ranjib works on end-to-end resiliency engineering practices across change, incident, and capacity management. Outside Uber, Ranjib is enthusiastic about Open Source and The Internet of Things.

Posted by Shu-Ming Peng, Jianing He, Ranjib Dey