This paper presents the information flow control model NetIFC to prevent information leakage when a net service is being executed. NetIFC offers the following features: (1) it blocks at least statements as possible and (2) it reduces runtime overhead. To achieve the first feature, NetIFC strictly controls output statements because only output may leak information. To achieve the second feature, NetIFC is executed in parallel with a service in different sites to monitor the service. This monitoring style substantially reduce runtime overhead when comparing with embedding a model in a net service.

Process-centered software engineering environments (PSEEs) facilitate managing software projects. According to the change of enactment environments and the increment of software development complexity, PSEE features should be enhanced. We designed a knowledge-based PSEE named KPSEE. It offers the features: (1) maximizing the degree of process parallelism, (2) enhancing process flexibility, (3) managing product consistency, (4) integrating PSEEs, (5) keeping pace with significant process change, (6) preventing technique leakage, and (7) offering project monitoring ability.

Information flow control models can be applied widely. This paper discusses only the models preventing information leakage during program execution. In the prevention, an information flow control model dynamically monitors statements that will cause information flows and ban statements that may cause leakage. We involved in the research of information flow control for years and identified that sensitive information may be leaked only when it is output. However, most existing models ignore information flows induced by output statements. We thus designed a new model XIFC (X information flow control) that especially emphasizes the monitoring of output statements. We also designed XIFC as a precise and low runtime overhead model. To achieve this purpose, we took a different viewpoint to re-examine the features offered by existing models and extracted a necessary feature set for the design. Our experiments show that XIFC bans every non-secure information flow and substantially reduces runtime overhead when comparing with our previous work.