The reputation of universities has drawn increasing attention in recent years, especially with the emergence of various rankings. However, despite advances in big data technologies that facilitate data collection and analysis, accurately defining and balancing factors related to university reputation and educational quality remains complex and tedious. Moreover, current educational assessment methods exhibit notable differences and controversies. In this paper, we present Iva , a human-in-the-loop I ntelligent V isual A ssessment system for higher education quality. This system utilizes large language models to analyze extensive multi-modal educational data, with visualization techniques incorporated to enable multi-scale exploration and interaction. Our extensive evaluations, including a carefully-designed user study and expert interviews, demonstrate the system’s potential value and provide insights for future improvements.

Load balancing in Remote Direct Memory Access (RDMA) networks is critical to network performance. Despite numerous load balancing algorithms devised to optimize multiple end-to-end links in data centers, they falter in RDMA networks due to unawareness of out-of-order packets and PFC pausing. Motivated by the above two issues, this paper proposes Gemma, a robust and path-aware load balancing algorithm. Gemma prevents out-of-order packets by proactively suspending and resuming the transmission of switch queues, thereby avoiding the performance degradation caused by retransmissions of out-of-order packets. In addition, Gemma leverages egress queue length as a path state indicator, integrating synchronized lengths from local and remote switches to comprehensively assess path congestion and facilitate prompt, effective rerouting to prevent frequently triggering PFC pausing. Moreover, Gemma can be extended to the Fat-Tree topology with a module that facilitates per-hop rerouting decisions on uplink paths, thereby enhancing scalability. Experiments show that compared to ConWeave, CONGA, and ECMP, Gemma achieves a 6%, 35%, and 57% improvement for average FCT under 80% load, which fully demonstrates the efficiency and feasibility of our proposed scheme.

In a typical distributed computing system, as the availability of resources and the precise execution time of different calculations are usually difficult to predict, how to effectively schedule complex calculations composed of interdependent tasks has become a challenge. The priority-based (PB) scheduling scheme, which is designed to maximize the parallelism of ready tasks, has shown better performance than other existing algorithms. However, the PB algorithm has the problem of excessive computing overhead and long running time in some cases. To address this issue, this paper proposes the priority-based level (PBL) algorithm. Experiments results show that the PBL , in comparison with their counterparts, manages to significantly reduce the algorithm running overhead while maintaining the scheduling results.

Abstract Nowadays, scientists are dealing with large‐scale scientific workflows that need a high processing capacity platform to facilitate on‐time completion. Cloud computing is the ideal platform to overcome this problem as it has several resources that scientists may choose from depending on the size of their applications. However, using cloud computing requires some monetary charges. Recently cloud computing providers started a new pricing schema that offers to their users a set of resources with specific combinations of CPU frequency configurations settings and price. The selected configurations settings reflect energy consumption. Besides, the configuration selection to meet users' satisfaction (minimum cost) and providers' satisfaction (energy saving) is crucial. Therefore, a multiobjective (cost and energy) efficient mechanism is essential. In this article, we address an important novel problem concerning multiobjective deadline constrained workflow scheduling in the cloud. We first study the relationship between cost minimization and minimization of the energy consumption in a cloud environment, and then discuss, develop, and propose an algorithm with two variants to help the system satisfy both sides (users and providers) at the same time during the selection of the configuration. The proposed heuristic is evaluated using specified real‐world applications. The observed results indicate that our heuristic can reduce significantly the energy consumption and the cost at the same time.

In this paper, we assume workflows under deadline constraints are submitted to the cloud from time to time. Every time a workflow is submitted, the cloud needs to determine whether it can agree with the specific constraint set by the user. If the cloud agrees to admit the workflow, cloud resources can be allocated for its execution in a way the deadline constraint can be met, while the existing load in the underlying resources is considered. The focus of this paper is how to schedule the tasks of each admitted workflow so that the resource utilization can be maximized. A variety of online scheduling algorithms have been proposed and evaluated using a simulator that manages to generate a stream of workflows for which an optimal schedule, with 100% resource utilization and without deadline violation, is guaranteed to exist.