Abstract:
Quality of service plays an important role in distributed systems. Cloud computing has been the recent entrant to the distributed system market and has revolutionized computing by changing the way computing resources are accessed and paid for. Users can access cloud services and pay only for the usage similar to utilities. Trust computing systems can be employed for identifying the service providers quality in terms of their adherence to the committed in SLAs. Several trust computing mechanisms have been proposed in literature based on various algorithms and functions. Almost all of them are based on a single performance parameter and modify the trust scores monotonously when performance deviations are reported. This paper proposes a trust computing mechanism that statistically validates the attribute monitored before modifying the trust scores. Hence the proposed mechanism is protected from momentary fluctuations in system performances. The experiments conducted show that the trust scores computed using the proposed mechanism are more representative of the long term system performance than the ones that were computed without the validation of the inputs.