Internet of Things (IoT) design focuses on concurrently handling multiple tasks for improving the scalability and robustness of the information sharing platform. Therefore, sophisticated resource allocation and optimization methods are necessary to prevent backlogs in request processing and resource allocation. This paper introduces a scalable resource allocation framework that is designed to maximize the service reliability in IoT because of a large volume of tasks and information. In this process, deep learning is used to assist the effective and scalable framework in allocating the resources to tasks with respective time constraints. The assisted allocation through deep learning balances the density of users, requests, and available resources without replications and overloading. Thus, the proposed deep learning based resource allocation framework helps in reducing the waiting and processing times of the requests under a controlled response time. Besides, the optimal segregation of available resources and request density facilitates failure-less allocation.

物联网（IoT）设计着重于并发处理多个任务，以提高信息共享平台的可扩展性和健壮性。因此，需要复杂的资源分配和优化方法来防止请求处理和资源分配中的积压。本文介绍了一种可扩展的资源分配框架，该框架旨在通过处理大量任务和信息来最大程度地提高IoT中的服务可靠性。在此过程中，深度学习用于协助有效且可扩展的框架将资源分配给具有相应时间限制的任务。通过深度学习进行的辅助分配可以平衡用户，请求和可用资源的密度，而不会造成复制和过载。从而，提出的基于深度学习的资源分配框架有助于在受控的响应时间下减少请求的等待和处理时间。此外，可用资源和请求密度的最佳隔离有助于实现无故障分配。