IMPROVE THE DATA ACCESS PERFORMANCE USING NEIGHBOR GROUP COOPERATIVE CACHING IN ADHOC NETWORKS

Mrs. Shanmugavadivu.K Professor , Department of M.C.A, K.S.Rangasamy College of Technology, Tiruchengode -637215 E-Mail: mahesh_ksv@rediffmail.com ABSTRACT Caching of frequently accessed data in multi-hop ad hoc environment is a potential technique that can improve the data access performance and availability. Cooperative caching, which allows the sharing and coordination of cached data among clients, can further explore the potential of the caching techniques. In this paper, we present a scheme, called Neighbor Group Cooperative (NGC) for caching in mobile ad hoc networks. In NGC scheme, multi-hop neighbors of a mobile client form a cooperative cache area since the cost for communication with them is low both in terms of energy consumption and message exchange. An analytical study of NGC based on data popularity, node density and transmission range is also performed. Simulation experiments show that the NGC caching mechanism achieves significant improvements in cache hit ratio and average query latency in comparison with other caching strategies. Keywords mobile networks, ad hoc networks, cooperative caching, cache consistency, cache placement. 1. INTRODUCTION Recent explosive growth in computer and wireless communication technologies has led to an increasing interest in mobile ad hoc networks (MANET) which are constructed only from Mobile hosts (MH). In ad hoc networks, every MH plays the role of a router for communication with other MHs. Even if the source and destination MH are not in communication range of each other, data packets are forwarded to the destination mobile host by relaying transmission through other MH which exist between the two MH [1]. However, in order to offer high quality and low cost services, several technical challenges still need to be addressed in ad hoc networks [2]. Most of the previous researches [15] in ad hoc networks focus on the development of dynamic routing protocols that can improve the connectivity among MHs. Although routing is an important issue in ad hoc networks, other issues such as data access are also very important since the ultimate goal of using such networks is to provide information access to MH [6]. Dr. Madheswaran .M The Principal, Muthayammal Engineering College, Rasipuram Caching has been proved to be an important technique for improving the data retrieval performance in mobile environments [1518]. With caching, the data access delay is reduced since requests can be served from the local cache, thereby obviating the need for data transmission over the scarce wireless links. However, caching techniques used in one hop mobile environment (i.e., cellular networks) may not be applicable to multi hop mobile environments since the data or request may need to go through multiple hops. As mobile clients in ad hoc networks may have similar tasks and share common interest, cooperative caching, which allows the sharing and coordination of cached data among multiple clients, can be used to reduce the bandwidth and power consumption. We investigate the data retrieval challenge of mobile ad hoc networks and propose a novel scheme, called Neighbor Group Cooperative (NGC) for caching. The goal of NGC is to reduce the cache discovery overhead and provide better cooperative caching performance. To enhance the system performance, individual caches interact with each other such that combined result is a larger cumulative cache. NGC dynamically chooses a super node as cache state node (CSN), to maintain the cache state (CS) information of different nodes within its domain. The CS for a client is the list of cached items along with their time-to-live (TTL) field. The rest of the paper is organized as follows. Section 2 reviews the related work. Section 3 describes the proposed CC caching scheme for data retrieval. Section 4 is devoted to performance evaluation and presents detailed simulation results. Section 5 concludes the paper. 2. RELATED WORK Caching is an important technique to enhance the performance of both wired and wireless network. A number of studies have been conducted to improve the caching performance in wireless mobile environment [1518]. Cooperative caching has been studied in the web environment, but little work has been done to efficiently manage the cache in ad hoc networks.

Journal of Computer Applications, Vol III, No.1, Jan - Mar 2010

Page 12

Due to mobility and constrained resources (i.e., bandwidth, battery power and computational capacity) in wireless networks, cooperative cache management techniques designed for wired networks may not be applicable to ad hoc networks. In the context of ad hoc networks, it is beneficial to cache frequently accessed data not only to reduce the average query latency but also to save wireless bandwidth. Hara [14] proposed several replica allocation methods to increase data accessibility and tolerate network partitions in MANETs. In these schemes, the replicated data are relocated periodically based on access frequency and overall network topology. Although replication can improve data accessibility, the overhead for relocating replicas periodically is significantly high. Due to updates at server, the cost of maintaining the consistent copy of replicas is quite high. Papadopouli and Schulzrinne [10] suggested the 7DS architecture, in which a couple of protocols re defined to share and disseminate information among users. It operates either on a prefetch mode, based on the information and users future needs or on an on-demand mode, which searches for data items in a one hop multicast basis. Depending on the collaborative behavior, a peer-to -peer (P2P) and server-to-client mode are used. Unlike our approach, this strategy focuses on data dissemination, and thus the cache management including cache admission control and replacement is not well explored. Sailhan and Issarny [16] proposed a cooperative caching scheme to increase data accessibility by P2P communication among MHs, when they are out of bound of a fixed infrastructure. It is implemented on top of Zone Routing Protocol (ZRP). The authors proposed a fixed broadcast range based on the underlying routing protocol. However, the mobile users location, data popularity and network density often change in a real mobile environment, so the fixed broadcast scheme is hard to adapt to real mobile applications. S. Lim et al. [17] proposed cache invalidation techniques for Internet based mobile ad hoc networks (IMANETs). However, due to broadcast nature, the cost of maintaining strong cache consistency is very high in ad hoc networks as compared to mobile networks. Das et al. [4] proposed a cooperative caching scheme for IMANETs. A broadcast based simple search scheme is proposed to establish cooperation among all MHs in the network to share cached data items. Although the broadcast based data search scheme can locate the nearest required data item, the energy and bandwidth cost of the flooding search is significantly high for a mobile ad hoc network. Shen et al. [08] proposed a broadcast based cooperative caching scheme for hybrid networks where a client shares the caches of clients lying in its proximity. Bandwidth and energy consumption to locate a client having cached the requested data

is very high due to flooding of the messages. Yin and Cao [6,7] proposed three caching algorithms to efficiently support data access in ad hoc networks. These algorithms mainly focus on the problem of choosing data item or data path for caching in the limited cache space of mobile nodes. Zang et al. [9] talk of security concerns for cooperative caching in ad hoc networks. 3. PROPOSED NEIGHBOR GROUP CACHING SCHEME A. Motivation In mobile ad hoc networks, MHs can move arbitrarily anytime and communicate with another one for data transmission. Although cooperative caching can provide the high access ability of data objects, the caching performance (cache hit ratio and average latency) can be reduced significantly due to the property of dynamic topology in MANETs. Possible reasons are listed as follows: 1. The caching nodes maybe shut down (leave) from the MANET because the energy of battery is exhausted. In other words, the cached data of caching nodes can not be serviced for others and will be removed after it re-connects the network. On the contrary, when an MH connects (joins) the network, its content of cache space is empty. If we can utilize its cache space as soon as possible, the caching performance will be enhanced. 2. The cache size of MHs always smaller than general personal computer. If there is no cooperative caching protocol among MHs, the cache space in all MHs can not be utilized effectively. If MHs can know the caching status of their neighbors, MHs can integrate their available cache space and efficiently cache data. 3. In MANETs, on-demand routing protocol is preferred to be used for saving energy and bandwidth. In this way, the caching node is selected based on the routing path between the source and the destination. Thus, if the multiple routing paths are passed through the same nodes, their cache space easily become full and the energy will be consumed fast. In order to deal with above conditions, we design a cooperative caching protocol among MHs. Our goal is to provide a caching and power efficient protocol in MANETs. First, an MH and its neighbors form a group. The group definition is presented in Section 3.2.2. Second, each MH sends their caching status to its group periodically. Third, when a data object needs to be placed or replaced in an MH, its group members cooperate to perform the tasks of placement and replacement. We assume MHs have

Journal of Computer Applications, Vol III, No.1, Jan - Mar 2010

Page 13

the ability to place the data in its group member [18]. The group caching has several benefits. First, it can reduce the redundancy of cached data object because MHs can check the caching status of other group members when receiving a data object. Second, it can store more different data objects and then increase the data accessibility. Third, the group can store more data objects from the destinations than an MH because the group members are cooperative to cache the data objects. B.Neighbor Group Caching (NGC) Network model: We model the MANET environment as a graph G = (V, E), where V represents the set of mobile hosts (MHs) in the network, and E is the set of links. An edge e = (u, v) E, where u, v V, exists if and only if u is in the transmission range of v and vice versa. All links in G are bi-directional,

the appropriate group member to place the cached data. Table I Simulation Parameters Parameter Default Value Range Database size (N) 1000 items smin 1 KB smax 10 KB Number of clients 70 50~100 (M) Client cache size 800 KB 200~1400 (C) KB Client speed 2 m/s 2~20 m/s (vmin~vmax) Bandwidth (b) 2 Mbps TTL 5000 sec 200~10000 sec Pause time 300 sec Mean query 5 sec 2~100 sec generate time (Tq) Transmission 250 m 25~250 m range (r) C. Placement and Replacement Policy In this section, we present how and where to place the data object in a group member when an MH receives a data object from the destination. The detailed algorithm is shown in Figure 2. Based on the usage of caching control message, each MH knows the remaining available cache space of other MHs in a group and the IDs and timestamps of their cached data objects.
Neighbor group Algorithm (di) If ( available cache space of receiving MH > size of di) Then Cache di Return Else If (available cache space of Neighbor MH > size of di) Then Push di to the neighbor MH randomly, Return Elseif If (available cache space of Next(Neighbor MH > size of di ) Then Push di to the Next(Neighbor)neighbor MH randomly, Return

Figure 1: The group in the view of MH C i.e., if u is in the transmission range of v, v is also in the transmission range of u. The network is assumed to be in a connected state. If it is partitioned, each component is treated as an independent network. Each MH and its one-hop neighbors form a group. A one-hop neighbor can be covered in the area of transmission range from an MH. Each MH has a group member ID. The group member ID may be the IP address or unique host ID. However, the biggest concern is the energy consumption in MHs and constrain of wireless bandwidth. Therefore, in our proposed GC scheme, each MH only maintains one-hop neighbors in a group. Figure 1 illustrates the group in the view of MH D. MH D and its one-hop neighbors {C, E, G and H} form a group. Thus, when a data item is received in MH D, MH D can check the caching status of each group member and select the appropriate group member to place the cached data. Thus, when a data item is received in MH D, MH D can check the caching status of each group member and select

Figure 2 : Pushing data item to a neighbor MH First of all, when an MH receives a data object (called receiving MH), it caches the data object if the cache space is enough. Otherwise, the receiving MH checks the available cache spaces of its Neighbor members. If the available cache space of any Neighbor member is sufficient to store the data object, the receiving MH puts the data object to the Neighbor member randomly Second, if the available cache space of every Neighbor is not sufficient to cache the received object, the receiving MH lookups the next neighbor of current neigbor. If the available cache space of every neighbor of current neighbor

Journal of Computer Applications, Vol III, No.1, Jan - Mar 2010

Page 14

member is sufficient to store the data object, the receiving MH puts the data object to the every neighbor of current neighbor member randomly. When a group member receives a data object from the receiving MH, it repeatedly performs the LRU replacement operations to increase the available cache space until the received data object can be cached. D.Data Discovery Process The process of data discovery performs the searches in the caching nodes for the requested object. In NeighborGroupCaching, when a requester (source) wants to retrieve a data object from the data source. If yes, it returns the data object (cache hit) to the application. If not, it lookups the next neighbor group member for retrieve a data object. Again if it not find the data object in the neighbor MH, it lookups the next near by neighbor for a data object. E. Cache Consistency Problem Two schemes can deal with the cache consistency problem: weak consistency and strong consistency. Under the weak consistency, a cached data object is associated with an attribute, TimeToLive (TTL). If the TTL time expires, the cached data object is removed. Under the strong consistency, if a cached data object is requested, the caching node first asks the data source to see if the cached data object is valid or not Because of the energy concern and the constrain of wireless bandwidth, we prefer using the weak consistency in mobile ad hoc networks. 4. Performance Evaluation The performance evaluation is shown in this section. The simulation model is given in Section 4.1. In Section 4.2, we verify the results of SimpleCaching, Cache path and CacheData [1],[6] and compared with the proposed NeighborGroupCaching. All schemes use LRU as the cache replacement policy. A. Simulation Results The simulation is performed on NS2 [9] with the CMU wireless extension. In our simulation, the AODV routing protocol [14] was tested as the underlying ad hoc routing algorithm. The data accessibility ratio increases with an increase in cache size for all caching schemes. For performance comparison with NGC, two other schemes Zone cooperative (ZC) caching and CacheData [6] are also implemented and query latency. The cache hit comprises of local hit, zone hit and remote hit. Fig. 3 shows the effect of cache size on the hit ratio For CacheData scheme, zone hit is always zero whereas zone hit and remote hit are zero for NC scheme. The local hit ratio for NC scheme is always the lowest. When cache size is small, CacheData performs similar to NC because they both use LRU replacement policy.

Fig. 4. Effect of mean query generate time on cache hit ratio The cache size is set to 200KB, 400KB, 600KB, 800KB, 1000KB, 1200KB and 1400KB. The size of a data item is set to 10KB. The pair of source and destination nodes is randomly selected in the simulation. In general, the cache hit ratio increases while the cache size increases. Fig 4 specifies the effect of mean query generate time on cache hit ratio. The Neighbor Group Caching has a higher cache hit ratio than others because both the MH and its group members can store data objects. These cached data objects improve the cache hit ratio.

Journal of Computer Applications, Vol III, No.1, Jan - Mar 2010

Page 15

REFERENCES [1] M. Frodigh, P. Johansson and L. Larsson, Wireless Ad Hoc Networking The Art of Networking Without a Network, Ericsson Review, No. 4, 2000. [2] S. Das, C. Perkins and E. Royer, Performance Comparison of Two On-Demand Routing Protocols for Ad Hoc Networks, IEEE INFOCOM,pp. 3-12, 2000. [3] D. Johnson and D. Maltz, Dynamic Source Routing in Ad Hoc WirelessNetworks, Mobile Computing, pp. 158-181, 1996. [4] C. Perkins and P. Bhagwat, Highly Dynamic Destination-SequencedDistance-Vector Routing (DSDV) for Mobile Computers, ACMSIGCOMM, pp. 234-244, 199. Fig. 5. The system performances as a function of the cache size Query delay. Figure 5 shows the experimental results under the dynamic topology. It describes the system performance of the cache size query delay. In every 20, 40, 60, 80, 100, and 120 seconds, ten MHs are selected randomly for joining and leaving the network. When an MH leaves the network, it removes all cached data objects. When the MH joins the network, the content of its cache is set to empty. NeighborGroupCaching shows the highest cache hit ratio because it utilizes all the available cache space of neighbors (group members). The average latency under different cache sizes and different join/leave rates. No communication with its neighbors is needed to know the caching status in other group members. As a result, the average latency is reduced compared with other schemes. 5. CONCLUSION In this paper, we propose a Neighborgroup caching scheme (NeighborGroupCaching) for mobile ad hoc networks. MHs maintain the localized caching status among the group members. Therefore, the MHs can cooperative to store different data objects. Furthermore, if an MH has available cache space, it can be utilized by its neighbors as soon as it joins a group. It improves the cache hit ratio and reduces the average latency compared with existing schemes. In future work we investigate the integration of broadcasting and cooperative caching. [5] C. Perkins and E.M. Royer, Ad Hoc OnDemand Distance Vector Routing, IEEE Workshop on Mobile Computing Systems and Applications, pp. 90-100, 1999. [6] L. Yin and G. Cao, Supporting Cooperative Caching in Ad Hoc Networks, IEEE INFOCOM, pp. 2537-2547, March 2004. [7] G. Cao, L. Yin and C. Das, Cooperative Cache Based Data Access Framework for Ad Hoc Networks, IEEE Computer, pp. 32-39, February 2004. [8] Huaping Shen, Sajal K. Das, Mohan Kumar and Zhijun Wang, Cooperative Caching with Optimal Radius in Hybrid Wireless Networks, NETWORKING, pp. 841-853, 2004. [9] W. Zhang, L. Yin and G. Cao, Secure Cooperative Cache Based Data Access in Ad Hoc Networks, NSF International Workshop on Theoretical and Algorithmic Aspects of Wireless Ad Hoc, Sensor, and Peer-to-Peer Networks, June 2004. [10] Maria Papadopouli and Henning Schulzrinne, Effects of Power Conservation, Wireless Coverage and Cooperation on Data Dissemination Among Mobile Devices, MobiHoc, October 2001. [11] Takahiro Hara, Replica Allocation Methods in Ad Hoc Networks with Data Update, Kluwer Journal of Mobile Networks and Applications,8(4), pp. 343-354, 2003. [12] N. Chand, R.C. Joshi and Manoj Misra, Broadcast Based Cache Invalidation and

Journal of Computer Applications, Vol III, No.1, Jan - Mar 2010

Page 16

Prefetching in Mobile Environment, International Conference on High Performance Computing (HiPC), Springer-Verlag LNCS 3296, pp. 410-419, 2004. [13] N. Chand, R.C. Joshi and Manoj Misra, Energy Efficient Cache Invalidation in a Disconnected Wireless Mobile Environment, International Journal of Ad Hoc and Ubiquitous Computing (IJAHUC), 2005. [14] G. Cao, On Improving the Performance of Cache Invalidation in Mobile Environments, ACM/Kluwer Mobile Networks and Applications, 7(4), pp. 291-303, 2002. [15] G. Cao, A Scalable Low-Latency Cache Invalidation Strategy for Mobile Environments, IEEE Transactions on Knowledge and Data Engineering, Vol. 15, No. 5, pp. 1251-1265, 2003. [16] F. Sailhan and V. Issarny, Cooperative Caching in Ad Hoc Networks, International Conference on Mobile Data Management (MDM), pp. 13- 28, 2003. [17] S. Lim, W.-C. Lee, G. Cao and C.R. Das, Performance Comparison ofCache Invalidation Strategies for Internet-Based Mobile Ad Hoc Networks, IEEE International Conference on Mobile Ad Hoc and Sensor Systems (MASS), pp. 104-113, October 2004. [18] Joonho Cho, Seungtaek Oh, Jaemyoung Kim, Hyeong Ho Lee, Joonwon Lee, Neighbor Caching in multi-hop wireless ad hoc netwoirk, IEEE Communication Letters, Volume &, Issue 11, Nov 2003 Page9s0 525-527 [19] Chand.N.,.Joshi. R.S. and Misra.M., Efficient Cooperative Caching in adhoc Networks communication system sofare and middleware , 2006, Comsware 2006. First International conference on 08-12 Jan. 2006 Page(s) :1-8 BIOGRAPHY Shanmugavadivu. K received the B.Sc Degree from Bharathiar university and MCA Degrees from Bharathidasan University and M.Phil (C.S) , M.Phil (C.S), Manonmaniam Sundaranar University. She is working as a Professor in Department of MCA at K.S.R College of Technology, Tiruchengode. Her area of interest Computer Networks, Network Protocols, Mobile Ad hoc Networks, Software Engineering. She is a member of ISTE.

M.Madheswaran received the BE Degree from Madurai Kamaraj University in 1990, ME Degree from Birla Institute of Technology, Mesra, Ranchi, India in 1992, both in Electronics and Communication Engineering. He obtained his PhD degree in Electronics Engineering from the Institute of Technology, Banaras Hindu University, Varanasi, India, in 1999. At present he is a Principal of Muthayammal Engineering College, Rasipuram, India. He has authored over forty five research publications in international and national journals and conferences. His areas of interest are theoretical modeling and simulation of high-speed semiconductor devices for integrated optoelectronics application, Bio-optics and Biosignal Processing. He was awarded the Young Scientist Fellowship (YSF) by the State Council for Science and Technology, TamilNadu, in 1994 and Senior Research Fellowship (SRF) by the Council of Scientific and Industrial Research (CSIR), Government of India in 1996. Also he has received YSF from SERC, Department of Science and Technology, Govt. of India. He is named in Marquis Whos Who in Science and engineering in the year 2006. He is a life member of IETE, ISTE and IE (India) and also a senior member of IEEE.

Journal of Computer Applications, Vol III, No.1, Jan - Mar 2010

Page 17