Results

• VANET bandwidth usage measures VANET bandwidth used for lookups performed in VANETs.

• P2P overlay bandwidth usage measures infrastructure network bandwidth used for lookups and overlay maintenance performed in a P2P overly.

3.7.2 Results

Figure 3.7 shows the lookup success rate under different number of vehicles. This figure shows that the single-tier VANET system has the lowest lookup success rate, especially in low density scenarios. This is because some lookups cannot reach the vehicles with the requested information in the disconnected VANETs. This problem can be alleviated by increasing the number of vehicles or introducing an infrastructure-based P2P overlay.

Both single-tier structured P2P and unstructured P2P systems significantly improve the lookup success rate because an infrastructure network does not have the disconnectivity problem and vehicles can communicate with any other vehicles through infrastructure-based communication. For the single-tier P2P systems, lookup success rate is independent of vehicle densities, and incorrect neighbor or finger information on P2P nodes is mainly caused by churn (i.e., node join/leave), resulting in lookup failures.

Simulation results show that the churn has an impact on the structured P2P approach.

Although the churn problem can be alleviated by frequently performing stabilization procedures to maintain a stable structured P2P overlay, these frequent stabilization procedures introduce extra maintenance overhead. On the other hand, the unstructured P2P approach adapts to churn much more effectively than the structured P2P approach in a dynamic vehicular network, as the single-tier unstructured P2P system achieves nearly a 95% lookup success rate.

38

The two-tier structured P2P system outperforms the single-tier VANET system because it also conducts lookups over the P2P overlay and can mitigate the disconnectivity problem of the VANETs. The two-tier structured P2P system has a success rate slightly lower than the single-tier structured P2P system because the periodical superpeer election may cause node join/leave in the P2P overlay and the structured P2P approach is vulnerable to churn.

By contrast, the two-tier unstructured P2P system is resilient to churn and achieves nearly a 95% lookup success rate, further accelerating the lookup and reducing maintenance costs compared to a single-tier unstructured P2P system, as will be shown later.

Figure 3.7: Lookup success rate for different approaches.

Figure 3.8 shows the average latencies of successful lookups. The single-tier VANET system achieves the shortest lookup latencies among all systems because of low-latency IVC. The lookup latency and success rate of the VANET system both increase with the

39

number of vehicles because more vehicles are in a connected VANET and a query must be propagated for more hops to reach the vehicle with the desired information.

The single-tier structured P2P system has the longest latency because the lookup hop count in the Chord-based overlay is proportional to the logarithm of the Chord network size. A successful lookup requires approximately six to eight hops in the Chord-based overlay as one-hop latency in infrastructure networks could be 600 ms long. Compared to the single-tier structured P2P system, the single-tier unstructured P2P system improves lookup latency by 50%–60%, because the geographic lookup in the Gnutella-based overlay can reach every vehicle within three hops.

By combining VANETs and an infrastructure-based P2P overlay, lookups can be simultaneously distributed over the VANETs and P2P overlay to improve the latency further. The two-tier systems outperform the single-tier P2P systems because low latencies can be achieved through lookups performed in VANETs. In summary, the two-tier systems achieve higher lookup success rate than the single-tier VANET system, and have lower lookup latency than the single-tier P2P systems. Moreover, the unstructured P2P approaches are more suitable for single-tier P2P and two-tier VANET/P2P systems because they achieve higher lookup success rates and introduce less lookup latencies than the structured P2P approaches.

40

Figure 3.8: Lookup latency for different approaches.

Figure 3.9 and Figure 3.10 shows the average bandwidth usage of VANETs (i.e., IVC lookup) and P2P overlay (i.e., P2P lookup and overlay maintenance in infrastructure network) per vehicle, respectively. The single-tier P2P systems do not consume VANET bandwidth as they only use infrastructure network. On the other hand, the single-tier VANET system utilizes only IVC, and thus does not consume the bandwidth of infrastructure network. As shown in Figure 3.9, the bandwidth usage of VANETs for the single-tier VANET and two-tier VANET/P2P systems significantly increases with the number of vehicles. The number of lookup messages increases in VANETs because more vehicles are involved in the lookups.

Figure 3.10 indicates the bandwidth usage of infrastructure network. The single-tier P2P and two-tier VANET/P2P systems occupy a certain infrastructure network bandwidth in performing the lookup and maintenance of the P2P overlay. The two-tier systems reduce

41

the bandwidth usage of infrastructure network by 40%–60% compared with the single-tier P2P systems because only some vehicles (i.e., superpeers) participate in the P2P overlay and perform the P2P operations. Although the unstructured P2P approach requires more bandwidth for P2P lookups, it still outperforms the structured P2P approach because of less P2P maintenance overhead.

Figure 3.9: VANET bandwidth usage for different approaches.

42

Figure 3.10: P2P overlay bandwidth usage for different approaches.

In the previous simulations, the superpeer election is performed every one second. The message overhead of maintaining the cluster structure decreases when the frequency of superpeer election decreases. When the election is performed every two seconds, the clustering message overhead in VANETs can be reduced while the lookup success rate is almost unchanged. However, if the superpeer election is performed much less frequently (e.g., longer than eight seconds), the lookup success rate decreases. This is because normal peers may move out of range of their superpeers and cannot communicate with their superpeers if the superpeer election is conducted infrequently. Although a vehicle can forward a lookup message to its VANET neighbors, the opportunity to send the lookup messages to the destination via P2P overlay may lose. On the other hand, the lookup latency is reduced slightly with a longer election period because less lookups are resolved

43

through P2P overlay which introduces more lookup latency than VANETs. Figure 3.11 and Figure 3.12 show the simulation results under different superpeer election frequencies, where Tclus indicates the clustering period (i.e., superpeer election period).

Figure 3.11: Lookup success rate under different clustering periods.

44

Figure 3.12: Lookup latency under different clustering periods.

The superpeer election is performed periodically to form clusters of vehicles within a range in VANETs. With one-hop clusters, vehicles are elected as superpeers if they have the largest IDs among one-hop neighbors. The one-hop clusters ensure that a normal peer can communicate directly with its superpeer to request a lookup in the P2P overlay.

However, because vehicles move and the VANET topology changes, normal peers may not be able to communicate with their superpeers through neighboring nodes especially for a larger cluster size, i.e., a larger number of hops for a cluster. The lookup success rate decreases when the number of hops for a cluster increases. Figure 3.13 and Figure 3.14 show the simulation results under different numbers of clustering hops (i.e., cluster size).

45

Figure 3.13: Lookup success rate under different clustering hops.

Figure 3.14: Lookup latency under different clustering hops.

46