International Business Machines Corporation

Fei Qi

Jing Li

Bin Xu

Guang Han Sui

Xue Ying Zhang

Yang Liang

Yue Liu

Na Liu

Yan Zhao Liu

Chen Yu Chang

Bin Shang

Wei Lei Wu

One or more computer processors extract respective features for each inter-modal sample in an inter-modal dataset, for each intra-modal sample in an intra-modal dataset, and a subsequent sample, wherein the inter-modal dataset and the intra-modal dataset are contained in a multi-modal training dataset. The one or more computer processors estimate an inter-modal label utilizing inter-modal label transformation of a subsequent sample. The one or more computer processors estimate an intra-modal label utilizing intra-modal label transformation of the subsequent sample. The one or more computer processors label the subsequent sample with a cross-modal label by combining the estimated inter-modal label and the estimated intra-modal label.

Cross-modal semi-supervised data labeling

A set of workload criteria is determined from a workload associated with a plurality of sources. The workload is divided among a set of workload groups according to the set of workload criteria and a first workload scheduler. A set of edge computing resources is assigned to each workload group within the set according to the set of workload criteria and the set of workload groups. A portion of the workload associated with a subset of the plurality of sources is handled by a first subset of edge computing resources and a second workload scheduler, where the subset of sources is associated with a first workload group. The handling includes balancing, by the second workload scheduler, the portion of the workload among the subset of sources. The handled workload is reported to a control center.

Edge computing workload balancing

Mechanisms are provided for managing caching of data in a multi-tenant cache storage device utilized by a distributed dataset based application. The mechanisms retrieve Resilient Distributed Dataset (RDD) block size information for an RDD associated with an application. A probability distribution of RDD block sizes is generated based on the RDD block size information and a maximum size for a RDD block of data is estimated based on the probability distribution. An amount of free space in a portion of the multi-tenant cache storage device allocated to the application is estimated based on the estimated maximum size for the RDD block of data. Cache operations for caching data associated with the application to the multi-tenant cache storage device are managed based on the estimated amount of free space in the portion of the multi-tenant cache storage device allocated to the application.

Caching a block of data in a multi-tenant cache storage device based on space usage boundary estimates

Growing community of inventors

Xi'an, China

Fei Qi