The growth of wireless communication and hardware design technologies boost the growth of mobile computing. Nowadays, mobile computing takes primary roles in various up-to-date computing environments such as ubiquitous and pervasive services. Infrastructure based communication strategies such as wireless cellular networks in which mobile devices are directed to fixed servers are mainly employed in mobile computing for providing stable services and supporting small-sized mobile devices.
As the number of mobile computing users and the types of mobile devices increase incredibly, the amount of information to process within the specified time is getting increase and ripple effects of information losses or server system failures are getting more critical. To provide satisfactory mobile computing services, servers for supporting mobile devices should 1) update and retrieve location of mobile devices efficiently, 2) provide fault-tolerant dependable service, and 3) disseminate common interested information to multiple mobile devices effectively. This thesis presents three approaches for these requirements employing hardware and software replication methods.
A default server strategy is commonly used to manage the location and state of mobile hosts in cellular networks. With this strategy, connections can be established after the client obtains the location information of the mobile host by querying the default server. However, the communication cost increases if the query requests are frequent and the distance between the default server and the client is long. Still more, any connection to a mobile host cannot be established when the default server of the destination mobile host fails. These problems can be solved by replicating default server and by letting the nearest replicated default server process the query request which is sent from a client. It is important to allocate replicated default servers efficiently in networks and determine the number of replicated default servers. This thesis suggests and evaluates a default server replication strategy to reduce communication costs and to improve service availabilities. Optimal replication degree and location for replicating default servers in n-grid and tree networks are also considered.
In ubiquitous environments, reasons of failures are complex and impacts of those are catastrophic for high dependable computing. In the hot-standby replication system, the system cannot process its tasks anymore when all replicated nodes have failed. Thus, the remaining living nodes should be well-protected against failure when parts of replicated nodes have failed. Design faults and system-specific weaknesses may cause chain reactions of common faults on identical replicated nodes in replication systems. These can be alleviated by replicating diverse hardware and software. Going one-step forward, failures on the remaining nodes can be suppressed by predicting and preventing the same fault when it has occurred on a replicated node in hot-standby replication system. This thesis analyzes system dependability by replicating diverse hardware and software for high available clustering computing. This thesis also proposes a fault avoidance scheme which increases system dependability by avoiding common faults on remaining nodes when parts of nodes have failed. It is verified that the proposed failure avoidance schemes can improve performance in three case studies.
Broadcasting mechanisms have been widely used to transfer information to a large number of clients. Information is transferred by broadcast servers (satellites or base stations) downstream with a wide bandwidth. Most of the broadcast schemes try to minimize the average “access time”. This thesis presents a real-time broadcast algorithm that transfers many information items including one with timing constraint. The proposed real-time broadcast algorithm attempts to meet the deadline for real-time information as well as to minimize the average access time for non real-time information. Simulation results show that the proposed algorithm can reduce the average access time for the non real-time information while meeting the deadline for the real-time information.