The Stanford Parallel Applications for Shared-Memory (SPLASH) is a set of parallel applications designed to support the design and evaluation of shared-memory multiprocessor systems. The goal is to provide a realistic and well-documented basis for evaluation studies, addressing the lack of real parallel applications available for system design. The applications cover various scientific and engineering domains, including scientific computations and computer-aided design. Each application is described in detail, including its problem domain, data structures, parallel structure, and profiling information. The evaluation section discusses performance results from both a real multiprocessor (Encore Multimax) and a simulator of an idealized architecture, highlighting the trade-offs between data locality, synchronization, granularity, and scalability. The paper also addresses the limitations of the applications, such as the impact of input data size and the need for restructuring on larger multiprocessors. The SPLASH suite is intended to evolve over time, with new applications added to enhance its coverage and utility for the parallel processing community.The Stanford Parallel Applications for Shared-Memory (SPLASH) is a set of parallel applications designed to support the design and evaluation of shared-memory multiprocessor systems. The goal is to provide a realistic and well-documented basis for evaluation studies, addressing the lack of real parallel applications available for system design. The applications cover various scientific and engineering domains, including scientific computations and computer-aided design. Each application is described in detail, including its problem domain, data structures, parallel structure, and profiling information. The evaluation section discusses performance results from both a real multiprocessor (Encore Multimax) and a simulator of an idealized architecture, highlighting the trade-offs between data locality, synchronization, granularity, and scalability. The paper also addresses the limitations of the applications, such as the impact of input data size and the need for restructuring on larger multiprocessors. The SPLASH suite is intended to evolve over time, with new applications added to enhance its coverage and utility for the parallel processing community.