Cost-effective Cloud Edge Traffic Engineering with Cascara
summary
Paper proposes Cascara, an online cloud-edge traffic-engineering controller that uses IPFIX flow records, BMP route data, and peering rates to solve a mixed-integer linear program approximating non-linear 95th-percentile billing. Evaluated on production demands of a commercial cloud, offline optimization shows ≈55% cost reduction potential and Cascara achieves 11–50% savings per PoP while restricting added client latency to about 3 ms by shifting traffic among latency-equivalent peer links.
AI-written summary · about
chapters (6)
- 0:00Start
- 0:34Inter-domain bandwidth costs
- 3:22Optimizing cloud bandwidth costs is hard
- 4:26Empirical insights can reduce problem scale
- 6:35Computing the best case cost savings
- 9:53Latency-equivalent links for outbound TE
related talks
18mCloudcast: High-Throughput, Cost-Aware Overlay Multicast in the Cloud
19mTackling DC Congestion and Bursts
17mBridging the Gap between QoE and QoS in Congestion Control: A Large-scale Mobile...
16mExpress Lane to Efficiency and Reliability: Multi-Dimensional Control in Meta’s Express..
13mEdgeBalance: Model-Based Load Balancing for Network Edge Data Planes
16mFLB: Fine-grained Load Balancing for Lossless Datacenter Networks
16mNetwork Flow Data in the Cloud
17mBarre: Empowering Simplified and Versatile Programmable Congestion Control in...
