Deep learning has achieved great success in numerous application areas at the cost of high computational complexity. To meet the ever-increasing computational demand, commodity hardware platforms (e.g., CPUs and GPUs) offer abundant computing resources including scalar, vector, and tensor units for deep learning that could execute in parallel. However, existing top-down tiling-based deep learning compilers often generate a homogeneous mapping from the given tensor computation task to hardware arithmetic instructions, failing to utilize different computing units simultaneously to achieve higher performance.

Big data processing clusters are suffering from a long job completion time due to the inefficient utilization of the RDMA capability. Our production measurement results in a large-scale cluster with hundreds of server nodes to process large-scale jobs have shown that the existing deployment of RDMA technique results in a long-tail job completion time, with some jobs even taking up more than twice the average time to complete. In this paper, we present the design and implementation of Turbo, an efficient communication framework for the large-scale data processing cluster to achieve high performance and scalability. The core of Turbo's approach is to leverage a dynamic block-level flowlet transmission mechanism and a non-blocking communication middleware to improve the network throughput and enhance system's scalability. Furthermore, Turbo ensures high system reliability by utilizing an external shuffle service as well as TCP serving as a backup. We integrate Turbo into Apache Spark and evaluate Turbo in a small-scale testbed and a large-scale cluster consisting of hundreds of server nodes. The small-scale testbed evaluation results show that Turbo improves the network throughput by 15.1% while maintaining high system reliability. The large-scale production results have shown Turbo can reduce the job completion time by 23.9% and increase the job completion rate by 2.03× over the existing RDMA solutions.

Now, autonomous tasks planning and allocating (TPA) in Multi Agent System (MAS) has been one key and fundamental problem to promote the intelligent level of such system. Autonomous TPA means that, all tasks should be (re)planned and (re)allocated automatically according to the synthesis constraints and the dynamic environment aspects, such as the changing mission, status of each member, and topology, etc. In this article, the formal descriptions of hierarchical tasks and models of logic constraints are studied firstly. And then, some new methods are proposed to evaluate the efficiency of synthesis constraints. Moreover, the key elements, e.g. task allocation vector (TAV), are designed with the theory of genetic algorithm (GA), and a TPA problem can be mapped to the solving model of GA. Based on above, the crossover and mutation operators of GA are optimized with the domain knowledge to perfect the solving efficiency and quality while ensuring the randomicity of evolution. The simulation results show that the solving quality and velocity are improved with studied methods.