Abstract: There is provided a method and apparatus for vectorized resource scheduling using multi-dimensional hierarchical arrays with a uniform data type. According to embodiments, the hierarchy of resources may be represented as a multi-dimensional array of a uniform data type in a scheduling context. This data structure is known as a tensor. In various embodiments, a tensor may be used to model each of resource usage, requested resources, resource capabilities, tenant resource usage, etc. There is provided tensor scheduling operations (e.g. tensor scheduling algorithms) to make use of this resource model. Some embodiments may exploit specialized hardware, such as GPUs and tensor-processing units, in order to accelerate the tensor scheduling operations.

Abstract: There is provided a method and apparatus for vectorized resource scheduling using multi-dimensional hierarchical arrays with a uniform data type. According to embodiments, the hierarchy of resources may be represented as a multi-dimensional array of a uniform data type in a scheduling context. This data structure is known as a tensor. In various embodiments, a tensor may be used to model each of resource usage, requested resources, resource capabilities, tenant resource usage, etc. There is provided tensor scheduling operations (e.g. tensor scheduling algorithms) to make use of this resource model. Some embodiments may exploit specialized hardware, such as GPUs and tensor-processing units, in order to accelerate the tensor scheduling operations.

Abstract: The present disclosure is drawn to methods and systems for providing a representative image for a search result in a search result listing of a searching system. At least one search result based on a search query is received. Each search result has a plurality of images. For each search result, a similarity score for at least some of the images is computed, based on a similarity with the search query. Then, at least one of the images is selected, on the basis of the similarity score, and the selected image is assigned as the representative image of the search result. A number of modification steps may be performed on the search result prior to computing the similarity score. The search result and assigned representative image may also be presented in a search result listing.

Abstract: Duration information indicative of an amount of time taken by each of one or more tasks of a distributed compute phase of a distributed compute job in a distributed compute cluster to execute is obtained. The one or more tasks are sorted into one or more groups based on the duration information and a resource requirement is determined for each of the one or more groups. A time-varying allocation of resources of the distributed compute cluster for the phase is determined based on the resource requirement for each of the one or more groups.

Abstract: A first quality of service identifier is assigned to each subtask associated with each node of a compute workflow, the first quality of service identifier indicative of a level of quality of service associated with each node. A planned resource requirement is assigned to each subtask, the planned resource requirement indicative of a total amount of system resources required to complete each subtask. A resource allocation plan is generated for each subtask, the resource allocation plan indicative of a distribution of the system resources over time in at least one resource manager. The resource allocation plan and the first quality of service identifier are output to the at least one resource manager for enforcement of the level of quality of service on one or more jobs submitted for each node through at least one workflow orchestrator external to the at least one resource manager.

Abstract: A system and method for subject-matching in images comprises detecting a first orientation of a first subject in a target image; comparing the first orientation with a second orientation of a second subject in a source image to obtain at least one orientation parameter; extracting at least one features from the target image; computing a visual similarity score between the target image and the source image using the features, the visual similarity score being computed as a function of the at least one feature and the at least one orientation parameter; and determining a match between the first subject and the second subject in accordance with the visual similarity score.

Abstract: Duration information indicative of an amount of time taken by each of one or more tasks of a distributed compute phase of a distributed compute job in a distributed compute cluster to execute is obtained. The one or more tasks are sorted into one or more groups based on the duration information and a resource requirement is determined for each of the one or more groups. A time-varying allocation of resources of the distributed compute cluster for the phase is determined based on the resource requirement for each of the one or more groups.

Abstract: A system and method for visualization of a compute workflow are provided. Workflow data indicative of one or more jobs executed in each computation stage of each instance of the compute workflow and job data indicative of one or more tasks executed by each of the one or more jobs are obtained. Each of the one or more tasks is associated with a corresponding computation stage. For each instance of the compute workflow, a visual representation of a resource utilization over time of each of the one or more tasks executed in each computation stage is generated. The visual representation is rendered for multiple instances of the compute workflow on a common timeline.

Abstract: A system and method for visualization of a compute workflow are provided. Workflow data indicative of one or more jobs executed in each computation stage of each instance of the compute workflow and job data indicative of one or more tasks executed by each of the one or more jobs are obtained. Each of the one or more tasks is associated with a corresponding computation stage. For each instance of the compute workflow, a visual representation of a resource utilization over time of each of the one or more tasks executed in each computation stage is generated. The visual representation is rendered for multiple instances of the compute workflow on a common timeline.

Abstract: A first quality of service identifier is assigned to each subtask associated with each node of a compute workflow, the first quality of service identifier indicative of a level of quality of service associated with each node. A planned resource requirement is assigned to each subtask, the planned resource requirement indicative of a total amount of system resources required to complete each subtask. A resource allocation plan is generated for each subtask, the resource allocation plan indicative of a distribution of the system resources over time in at least one resource manager. The resource allocation plan and the first quality of service identifier are output to the at least one resource manager for enforcement of the level of quality of service on one or more jobs submitted for each node through at least one workflow orchestrator external to the at least one resource manager.

Abstract: The present disclosure is drawn to methods and systems for providing a representative image for a search result in a search result listing of a searching system. At least one search result based on a search query is received. Each search result has a plurality of images. For each search result, a similarity score for at least some of the images is computed, based on a similarity with the search query. Then, at least one of the images is selected, on the basis of the similarity score, and the selected image is assigned as the representative image of the search result. A number of modification steps may be performed on the search result prior to computing the similarity score. The search result and assigned representative image may also be presented in a search result listing.

Abstract: A system and method for subject-matching in images comprises detecting a first orientation of a first subject in a target image; comparing the first orientation with a second orientation of a second subject in a source image to obtain at least one orientation parameter; extracting at least one features from the target image; computing a visual similarity score between the target image and the source image using the features, the visual similarity score being computed as a function of the at least one feature and the at least one orientation parameter; and determining a match between the first subject and the second subject in accordance with the visual similarity score.