This paper presents an overview of distributed discrete-event simulation (DDES), focusing on its theory and challenges. Traditional discrete-event simulations are inherently sequential, limiting their ability to simulate large systems due to the modest number of events they can handle. DDES, which runs on a network of processors with asynchronous communication, offers a potential alternative by partitioning the simulation among component processors, potentially improving performance. The basic DDES scheme uses time encoding, but it may result in deadlock. The paper discusses various techniques for deadlock avoidance and detection, emphasizing the theory behind DDES. The paper begins by introducing the problem of system simulation, which typically involves sequential steps: fetching an event, simulating it, and possibly updating the data structure. Traditional simulation algorithms are not easily parallelizable due to the sequential nature of the event-list mechanism. DDES, however, allows for concurrent execution by distributing the simulation across multiple machines, with each machine simulating a single physical process. Messages between processes are used to simulate interactions, and time is encoded in messages to capture the synchronous nature of the physical system. The paper discusses the limitations of sequential simulation, such as the inability to handle large systems due to the sheer magnitude of the problem. It then introduces the concept of distributed simulation, which requires little additional memory and allows for simulation to be adapted to the structure of available hardware. The paper also highlights the importance of deadlock resolution in DDES, as deadlock can occur when processes are waiting for messages that never arrive. The paper presents a basic distributed simulation scheme, which is shown to be partially correct and may result in deadlock. Various approaches for deadlock resolution are discussed, including the use of null messages and circulating markers. The paper concludes that while the basic theory of DDES has been developed, further research is needed to ensure substantial performance gains over sequential simulation. The paper also notes that DDES can be implemented using existing simulation languages and that it avoids traditional issues such as pseudorandom number generation. The paper emphasizes the importance of modeling interactions through message communication and the need for further study to improve the performance of DDES.This paper presents an overview of distributed discrete-event simulation (DDES), focusing on its theory and challenges. Traditional discrete-event simulations are inherently sequential, limiting their ability to simulate large systems due to the modest number of events they can handle. DDES, which runs on a network of processors with asynchronous communication, offers a potential alternative by partitioning the simulation among component processors, potentially improving performance. The basic DDES scheme uses time encoding, but it may result in deadlock. The paper discusses various techniques for deadlock avoidance and detection, emphasizing the theory behind DDES. The paper begins by introducing the problem of system simulation, which typically involves sequential steps: fetching an event, simulating it, and possibly updating the data structure. Traditional simulation algorithms are not easily parallelizable due to the sequential nature of the event-list mechanism. DDES, however, allows for concurrent execution by distributing the simulation across multiple machines, with each machine simulating a single physical process. Messages between processes are used to simulate interactions, and time is encoded in messages to capture the synchronous nature of the physical system. The paper discusses the limitations of sequential simulation, such as the inability to handle large systems due to the sheer magnitude of the problem. It then introduces the concept of distributed simulation, which requires little additional memory and allows for simulation to be adapted to the structure of available hardware. The paper also highlights the importance of deadlock resolution in DDES, as deadlock can occur when processes are waiting for messages that never arrive. The paper presents a basic distributed simulation scheme, which is shown to be partially correct and may result in deadlock. Various approaches for deadlock resolution are discussed, including the use of null messages and circulating markers. The paper concludes that while the basic theory of DDES has been developed, further research is needed to ensure substantial performance gains over sequential simulation. The paper also notes that DDES can be implemented using existing simulation languages and that it avoids traditional issues such as pseudorandom number generation. The paper emphasizes the importance of modeling interactions through message communication and the need for further study to improve the performance of DDES.