Testing static random access memories (SRAM's) for all possible failures is not feasible. We have to restrict the class of faults to he considered. This restricted class is called a fault model. A fault model for SRAM's is presented based on physical spot defects, which are modeled as local disturbances in the layout of an SRAM. Two linear test algorithms are proposed, that cover 100% of the faults under the fault model. A general solution is given for testing word oriented SRAM's. The practical validity of the fault model and the two test algorithms are verified by a large number of actual wafer tests and device failure analyses.

No abstract

Summary Structures are subjected to cyclic loads that can vary in direction and magnitude, causing constant amplitude mode I simulations to be too simplistic. This study presents a new approach for fatigue crack propagation in ductile materials that can capture mixed‐mode loading and overloading. The extended finite element method is used to deal with arbitrary crack paths. Furthermore, adaptive meshing is applied to minimize computation time. A fracture process zone ahead of the physical crack tip is represented by means of cohesive tractions from which the energy release rate, and thus the stress intensity factor can be extracted for an elastic‐plastic material. The approach is therefore compatible with the Paris equation, which is an empirical relation to compute the fatigue crack growth rate. Two different models to compute the cohesive tractions are compared. First, a cohesive zone model with a static cohesive law is used. The second model is based on the interfacial thick level set method in which tractions follow from a given damage profile. Both models show good agreement with a mode I analytical relation and a mixed‐mode experiment. Furthermore, it is shown that the presented models can capture crack growth retardation as a result of an overload.

No abstract

We present a framework for recognition of data structuresin programs, to aid in design recovery. Theframework consists of an intermediate representation anda knowledge base containing information about typical implementations of abstract data types. The framework is suited for recognition of data structures combined with their characteristic operations. Abstract data structures can be recognized partially, they can be wcognized even i f they are delocalized, and different independent interpretations of the same structures can be generated.