摘要:
A regenerated container file is detected, and a file in the regenerated container file is determined that is different from any file in an existing container file related to the regenerated container file. To store the regenerated container file, the different file is sent to the data storage for storing.
摘要:
A user may store content in a data store that includes multiple repositories corresponding to one or more categories, where each of the repositories may include one or more containers for content storage. An inference application may be configured to automatically save the content to a location within the one or more containers of the repository. For example, the inference application may be configured to detect content to be saved for a user. The inference application may analyze one or more attributes of the user and the content to infer a repository category to which the content is related, and determine a confidence level associated with the inferred repository category. In response to a determination that the confidence level is above a threshold, the content may be automatically saved to a determined location within one or more containers of a repository corresponding to the inferred repository category.
摘要:
Performance optimization for reduced and bounded memory cost for data visualization is provided. Performance optimization comprises: data culling, geometry culling, and cloning of a visualization to a background thread for layout. The performance optimization leverages a data visualization architecture for building of a data visualization via a one-directional chain of separate stages, wherein data at each stage may be culled or privatized to reduce the amount of data, or simplify the nature of the data, to be processed in subsequent stages, thus improving overall system performance and user experience.
摘要:
Systems and methods for generating a grammar describing activities of a user are disclosed. An aspect receives log data for the user, clusters the log data around a plurality of cluster centroids, assigns one or more semantic labels to each of the plurality of cluster centroids based on determining that a threshold number of log data points have been assigned to each of the plurality of cluster centroids, determines a sequence in which the log data points were clustered around the plurality of cluster centroids, generates one or more grammars representing a sequence of possible activities of the user based on the sequence in which the log data points were clustered around the plurality of cluster centroids and the one or more semantic labels of each of the plurality of cluster centroids, and filters the assigned one or more semantic labels for each of the plurality of cluster centroids.
摘要:
Methods, computer-readable media, and systems for scheduling the association of metadata with content. In an embodiment, event information is obtained from a virtual calendar, wherein the event information comprises at least one event detail and one or more parameters defining a time period. First metadata is generated based on the event detail, and is stored, in association with the time period, in a memory. Then, subsequently, during the time period, the first metadata may be retrieved from the memory, and associated with one or more content items generated on the device.
摘要:
A single system merges primary data storage, data protection, and intelligence. Intelligence is provided through in-line data analytics, and data intelligence and analytics are gathered on protected data and prior analytics, and stored in discovery points, all without impacting performance of primary storage. Real-time analysis is done in-line with the HA processing, enabling a variety of data analytics that are then used as part of a live restore operation. Data content can be live restored at an object or block level. Data recovery begins with metadata restoration, followed by near-instantaneous access to "hot" regions of data being restored, allowing site operation to continue or resume while a restore is ongoing.
摘要:
본 발명은 대용량의 CFD 데이터를 병렬로 처리하여 가시화 하는데 있어서 대용량 데이터를 구조 격자 또는 비구조 격자의 데이터 파일 포맷 구조로 생성하여 저장하거나 기록하여 둘 수 있도록 하는, 대용량 CFD 병렬 데이터의 가시화를 위한 데이터 파일 포맷 구조로 데이터가 기록된 기록 매체 및 그 데이터 파일 포맷 구조 생성 방법에 관한 것이다. 본 발명의 일 실시예에 따른 대용량 CFD 병렬 데이터의 가시화를 위한 구조 격자(Structured Grid)의 데이터 파일 포맷 구조로 데이터가 기록된 기록 매체에 있어서, 상기 구조 격자의 데이터 파일 포맷 구조는, 메쉬(mesh)와 데이터(data)에 대한 엘리먼트(element)의 수와 ID, 타임스텝(time step)의 수와 ID, 그리고 상기 데이터(data)의 밸류(value)의 수와 ID를 포함하는 전체 데이터에 대한 특성 정보를 기술하는 메타(meta) 데이터부; 상기 메쉬에 대해 엘리먼트(element) 단위의 디렉토리(directory)와 상기 엘리먼트 내에 타임스텝(time step)별 디렉토리가 격자 구조로 구조화 된 메쉬 디렉토리부; 및 상기 데이터에 대해 엘리먼트(element) 단위의 디렉토리(directory)와 상기 엘리먼트 내에 타임스텝(time step)별 디렉토리가 격자 구조로 구조화되고, 상기 타임스텝별 디렉토리마다 하나 이상의 밸류(value)별 디렉토리(directory)가 구조화 된 데이터 디렉토리부를 포함한다.
摘要:
A file system is configured to provide, based on a request from a requester, a requested file stored on a storage medium to the requester. Furthermore, the file system is configured to provide a chosen part of the requested file corresponding to a desired version of the content of the requested file to the requester, based on received side information in the request indicating the desired version, and based on meta information of the requested file.
摘要:
The performance of the analysis system is deteriorated because file content extraction processing is performed in the file aggregation server and in the analysis server and further because annotation data creation is performed in the file aggregation server. Therefore, the present invention solves the problem by providing a file aggregation server classifying files into analysis target contents, non analysis target contents, and content matched data, and providing only the analysis target contents to the analysis server. Since this method enables the analysis server to acquire the analysis target contents directly from the file aggregation server, the processing of extracting contents from the files becomes unnecessary, and the throughput of the entire analysis system is improved.