فهرست مطالب وحید شاهی

In the livestock industry, the accurate genetic evaluation of economic traits pre-request the existence of homogeneity in residual variance, however; in practice, such different production levels and data recording at different times and environments are unlikely to result in the homogeneity of residual variance. The aim of this study was to investigate the genetic variance of the residual variance for Egg Weight and Age at Sexual Maturity of Azerbaijan native chickens. Data set consisted 8283 Egg Weight (EW) records and 6608 Age at Sexual Maturity (ASM) records related to 9 generations from the indigenous chicken breeding centers in the west Azerbaijan. To investigate the genetic variation of the residual variance was using the BLUPF90 software. Heritability of EW and ASM in a single trait model was estimated to be 0.288 (s.e. =0.022) and 0.355 (s.e =0.127), respectively. The heritability of residual variance of these traits was estimated 0.00484 (s.e. =0.004) and 0.00005 (s.e. =0.002), respectively, indicating the slight possibility of their genetic progression by direct genetic selection. The values of genetic correlations between traits and their residual variance were estimated to be -0.154 and -0.076, respectively, suggesting that genetic selection for EW enhancement would improve the genetic value for their uniformity, but about ASM, the expected results will not be very favorable if the trait is reduced. In conclusion, for accurate estimation of breeding value and ranking based on production traits in poultry, special attention to variance heterogeneity and necessary corrections is inevitable.

Cloud computing is a processing model that has been of interest to researchers for reasons of flexibility, rapid deployment and low cost and also provides shared use of services in a network environment without the need to the ownership right and direct management of resource. Due to the storage and retrieving data of cloud services users, data security is one of the major challenges of the cloud computing. Several data security models in the cloud have been proposed to deal with these challenges, which we have evaluated some of them. The results of evaluations show that the combined data security model has the most security when the data in addition to encryption is also classified and indexed. Also the level- based model tries to secure the data in the cloud by data division into two levels and double authentication. The data security Business model with attention to the independent activity of cloud services, has also lower overhead than other data security models.