Abhigyan, Aditya Mishra, Vikas Kumar, Arun Venkataramani University of Massachusetts Amherst 1
Examples: ◦ CDNs ◦ P2P applications ◦ Mirrored websites ◦ Cloud computing 2 Location diversity: Ability to download content from multiple locations
ISPs have several objectives, e.g., minimizing congestion, decisions about upgrading link capacity ISPs optimize link utilization based metrics. e.g. maximum link utilization (MLU) 3
4 Traffic engineering (ISPs) Location diversity (CDNs) Internet traffic
How do TE schemes compare accounting for location diversity in the Internet? 5
1. Introduction 2. Motivation 1.Location diversity and traffic engineering 2.Metric of comparison 3. Evaluation 4. Conclusion 6
Application adaptation to location diversity Traffic matrix New Routing 7 Traffic engineering Content demand
8 100 Mbps, 0.1ms 100 Mbps, 10ms Mb x 10 req/s = 100 Mbps 10 Mb x 5 req/s = 50 Mbps OSPF Wt = 2 OSPF Wt = 1 50 Mbps + 50 Mbps 50 Mbps Maximum link utilization ( MLU )= 1 OSPF Wt = 1
9 100 Mbps, 0.1ms 100 Mbps, 10ms OSPF Wt = 2 OSPF Wt = 1 50 Mbps + 50 Mbps 50 Mbps OSPF Wt = 1 25 Mbps + 25 Mbps Expected MLU = Mbps +25Mbps MLU = Mb x 10 req/s = 100 Mbps 10 Mb x 5 req/s = 50 Mbps OSPF Wt = 1
Location diversity increases capacity 100 Mbps Mb x 10req/s = 100 Mbps 100 Mbps 10 Mb x 20req/s = 200 Mbps Increase in capacity = 200/ 100 = 2
1. Motivation 1.Location diversity and traffic engineering 2.Metric of comparison 2. Evaluation 3. Conclusion 11
Without location diversity ◦ Capacity = 1/MLU Mbps Mbps 100 Mbps MLU = Mbps Capacity = 100/25 = Mbps max supportable demand current demand Capacity =
Without location diversity ◦ Capacity = 1/MLU With location diversity ◦ Ca pacity >= 1/MLU Mbps Mbps 100 Mbps 25 Mbps 5 Mbps MLU = Mbps 90 Mbps Capacity > 180/30 = 6 Need a new metric to quantify capacity under location diversity max supportable demand current demand Capacity =
SPF = Maximum supportable surge (linearly scaled) in traffic demand 14 SPF = 200/30 = Mbps Mbps 100 Mbps 25 Mbps 5 Mbps 100 Mbps 200Mbps
Location diversity significantly impacts TE 1.Capacity increases 2.Capacity (SPF) not captured by 1/MLU 15
1. Introduction 2. Motivation 3. Evaluation 1.TE schemes 2.Measuring SPF 3.Capacity results (SPF) 4. Conclusion 16
17 TE Schemes (Almost online) optimal TE [OPT] (Offline) “optimal” TE using MPLS [MPLS] (Offline) TE using OSPF link weight optimization [OptWt] (Offline) Multi-TM optimization TE [COPE] (Oblivious) Static shortest path routing with inverse- capacity link weights [InvCap]
18 Is demand satisfied ? Increase demand by Δ SPF = demand/(initial demand) Demand = initial demand YES NO
19 InvCap worst case No LocDiv = 50% sub-OPT LocDiv = 30% sub-OPT 1.All TE schemes achieve near-optimal capacity with location diversity. 2.Even no TE scheme is at most 30% sub-optimal with location diversity.
“How location diversity ate traffic engineering’s cake” ◦ Any TE scheme performs the same as Optimal TE. ◦ No TE scheme performs at most 30% worse. 20