中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

7 records (took 0.015 seconds)

    Network service for identifying infrequently accessed data in a data stream
    3.
    发明授权
    Network service for identifying infrequently accessed data in a data stream 有权

    公开(公告)号：US10523756B1

    公开(公告)日：2019-12-31

    申请号：US15446774

    申请日：2017-03-01

    Applicant: Amazon Technologies, Inc.

    Inventor： Anand Chakraborty ,  Michael Christopher Wenneman

    IPC: G06F15/173 ,  H04L29/08 ,  G06F3/06 ,  H04L29/06 ,  G06N20/00

    Abstract: A stream of storage requests can be received for data objects stored by a storage service. A streaming algorithm can be utilized to identify the most frequently accessed objects stored by the storage service. A statistical distribution of the most frequently accessed objects can then be generated and utilized to estimate the number of infrequently accessed objects. Machine learning can also be utilized to identify correlations between attributes of objects stored by the storage service and their associated access rates. For instance, machine learning can be utilized to determine that objects stored in a certain location or having other characteristics typically have low access rates. Information regarding the number of infrequently accessed objects and their learned attributes can be utilized to take action with regard to the infrequently accessed objects, such as moving the infrequently accessed objects to long-term storage.

    Analysis of interactions with data objects stored by a network-based storage service
    5.
    发明授权
    Analysis of interactions with data objects stored by a network-based storage service 有权

    公开(公告)号：US11481662B1

    公开(公告)日：2022-10-25

    申请号：US15664765

    申请日：2017-07-31

    Applicant: Amazon Technologies, Inc.

    Inventor： Anand Chakraborty ,  Thayn Moore ,  Bhavesh Anil Doshi

    IPC: G06N20/00

    Abstract: Technologies are described for analyzing interactions with data objects stored by a network-based storage service. The analysis of the interactions can identify patterns of the data object interactions and outcomes that can result from the patterns. Models can be developed that include the patterns and the outcomes corresponding to the patterns. As requests related to data object interactions are subsequently obtained by the system, the requests can be analyzed with respect to the models to identify an outcome that may be associated with the requests.

Patent Agency Ranking