Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for generating images. The method includes acquiring a descriptive text for describing image content of a target image, determining position prior information, and generating the target image based on the descriptive text and position prior information. According to the method of the embodiments of the present disclosure, a target image that can be used for rare data simulation in rare scenarios can be generated by means of an input descriptive text. In addition, the method allows for position perception editing and operation, and can control, based on position prior information, a direction and a position of an object generated in the target image, thereby effectively and diversely generating images. Moreover, the method provided in the present disclosure is based on object types in each subdivided image block, thus making position perception more accurate.

Abstract: The present disclosure relates to de-duplication. An example method includes acquiring mirror files, wherein each mirror file in the mirror files includes at least one layer. The method further includes generating a directory by de-duplicating layers of the mirror files. The method further includes redirecting the de-duplicated mirror files to the directory by using hard links. The method further includes packaging the redirected mirror files and the directory into a target package. Beneficially, the mirror files can be de-duplicated in a unit of layer, and the flexibility in downloading a part of the mirror files can further be ensured.

Abstract: Techniques described herein relate to a method for storing data in columnar storage. The method includes obtaining a columnar storage write request associated with a file, wherein the file comprises rows and columns of file data; in response to obtaining the columnar storage write request: writing the file data to storage using column-based caches; generating file metadata based on the writing of the file data to the storage; and assigning a key to the file metadata; and storing the file metadata using a key-value service.

Abstract: Data placement is adjusted. For instance, it is determined whether a node Failure To Tolerate (FTT) of a chunk in a storage system meets an optimized node FTT corresponding to the chunk, and, when it is determined that it does not meet, an FTT task entry for the chunk is added to a chunk index of the storage system. The added FTT task entry includes an identifier of the chunk, its node FTT, and the corresponding optimized system node FTT. Furthermore, instead of scanning placement information for each chunk, the storage system may adjust, based on the FTT task entry in the chunk index, placement of blocks of the corresponding chunk in system nodes. Therefore, the need to scan the placement information of all chunks and calculate their node FTTs in rebalance operations and minimum node FTT queries is eliminated, thereby improving the performance of the storage system.

Abstract: The technologies described herein are generally directed toward writing multiple data objects to object storage. An example method can include receiving an aggregated data object comprising a set of data objects and metadata corresponding to respective data objects of the set of data objects. The method can further include updating the metadata to comprise the respective storage locations. Further, the method can include updating the metadata to include the respective storage locations. The method can further include allocating the metadata to be stored by metadata processing equipment, resulting in the metadata being stored in metadata storage equipment.

Abstract: Data placement is adjusted. For instance, it is determined whether a node Failure To Tolerate (FTT) of a chunk in a storage system meets an optimized node FTT corresponding to the chunk, and, when it is determined that it does not meet, an FTT task entry for the chunk is added to a chunk index of the storage system. The added FTT task entry includes an identifier of the chunk, its node FTT, and the corresponding optimized system node FTT. Furthermore, instead of scanning placement information for each chunk, the storage system may adjust, based on the FTT task entry in the chunk index, placement of blocks of the corresponding chunk in system nodes. Therefore, the need to scan the placement information of all chunks and calculate their node FTTs in rebalance operations and minimum node FTT queries is eliminated, thereby improving the performance of the storage system.

Abstract: The technologies described herein are generally directed toward writing multiple data objects to object storage. An example method can include receiving an aggregated data object comprising a set of data objects and metadata corresponding to respective data objects of the set of data objects. The method can further include updating the metadata to comprise the respective storage locations. Further, the method can include updating the metadata to include the respective storage locations. The method can further include allocating the metadata to be stored by metadata processing equipment, resulting in the metadata being stored in metadata storage equipment.

Abstract: The technologies described herein are generally directed toward retrieving data from streaming storage. In an embodiment, a method can include receiving an application data request that identifies application data to be retrieved from a sequence of stored data chunks that correspond to a stored stream of data. The method can further include, based on the application data request, estimating a first estimated location of the application data, with the first estimated location including an identified chunk of a sequence of chunks. Further, the method can include, based on the application data request and a characteristic of the identified chunk, retrieving, by the system, a first data block that is estimated to comprise the application data, resulting in a first retrieved data block.

Abstract: The technologies described herein are generally directed toward retrieving data from streaming storage. In an embodiment, a method can include receiving an application data request that identifies application data to be retrieved from a sequence of stored data chunks that correspond to a stored stream of data. The method can further include, based on the application data request, estimating a first estimated location of the application data, with the first estimated location including an identified chunk of a sequence of chunks. Further, the method can include, based on the application data request and a characteristic of the identified chunk, retrieving, by the system, a first data block that is estimated to comprise the application data, resulting in a first retrieved data block.

Abstract: The technologies described herein are generally directed toward using chunks of data to store streaming data at a cloud service provider. In an embodiment, a method can include, based on first application data, initiating a cloud storage transaction, corresponding to a chunk of data. The method can further include communicating a first data part of the chunk to the cloud storage server. Further, the method can include, based on second application data, communicating a second data part of the chunk to the cloud storage server, with the first data part and the second data part being, as generated, stored in a commit buffer, and with a first chunk offset and a second chunk offset being stored in metadata. The method can further include, based on a size of the chunk, communicating a commit signal to commit the cloud storage transaction.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for generating images. The method includes acquiring a descriptive text for describing image content of a target image, determining position prior information, and generating the target image based on the descriptive text and position prior information. According to the method of the embodiments of the present disclosure, a target image that can be used for rare data simulation in rare scenarios can be generated by means of an input descriptive text. In addition, the method allows for position perception editing and operation, and can control, based on position prior information, a direction and a position of an object generated in the target image, thereby effectively and diversely generating images. Moreover, the method provided in the present disclosure is based on object types in each subdivided image block, thus making position perception more accurate.

Abstract: A system of managing data, comprising a streaming data storage system operating on a processor that causes the processor to receive streaming data and to process the streaming data in response to an append command, a long-term storage system operating on a processor that causes the processor to receive the processed streaming data from the streaming data storage system and to generate a chunk of data, an object storage system operating on a processor that causes the processor to receive the processed streaming data from the streaming data storage system and to generate an object of data and a tables system operating on a processor that causes the processor to receive the processed streaming data from the streaming data storage system and to generate a table of data.

Abstract: Techniques described herein relate to a method for storing data in columnar storage. The method includes obtaining a columnar storage write request associated with a file, wherein the file comprises rows and columns of file data; in response to obtaining the columnar storage write request: writing the file data to storage using column-based caches; generating file metadata based on the writing of the file data to the storage; and assigning a key to the file metadata; and storing the file metadata using a key-value service.

Abstract: A system of managing data, comprising a streaming data storage system operating on a processor that causes the processor to receive streaming data and to process the streaming data in response to an append command, a long-term storage system operating on a processor that causes the processor to receive the processed streaming data from the streaming data storage system and to generate a chunk of data, an object storage system operating on a processor that causes the processor to receive the processed streaming data from the streaming data storage system and to generate an object of data and a tables system operating on a processor that causes the processor to receive the processed streaming data from the streaming data storage system and to generate a table of data.

Abstract: Support for dynamic scaling is provided. An example method includes: acquiring a current value of a metric associated with a plurality of components of an application; determining a target number of replicas of at least one component of the plurality of components based on the current value of the metric and a scaling policy for the plurality of components; and updating a configuration manifest of the application based on the target number of replicas of the at least one component. In this manner, the configuration manifest of the application can be automatically updated based on metric values of the components at runtime and a custom scaling policy, thereby achieving dynamic scaling of various components of the application without additional user effort to implement separate dynamic scaling mechanisms.

Abstract: A system of managing data, comprising a streaming data storage system operating on a processor that causes the processor to receive streaming data and to process the streaming data in response to an append command, a long-term storage system operating on a processor that causes the processor to receive the processed streaming data from the streaming data storage system and to generate a chunk of data, an object storage system operating on a processor that causes the processor to receive the processed streaming data from the streaming data storage system and to generate an object of data and a tables system operating on a processor that causes the processor to receive the processed streaming data from the streaming data storage system and to generate a table of data.

Abstract: Indices of storage systems are managed. An example method includes: receiving a flush cycle for flushing expired events in a storage system including multiple events in a data stream, determining, based on the flush cycle, a time slice for managing the index of the storage system, creating a slice index node for the time slice in response to determining that the occurrence time of a first event that enters the storage system in the data stream is within the time slice, wherein the slice index node includes an index node of the first event, and adding the slice index node to the index. A corresponding device and a corresponding computer program product are provided. Thus, the index of a large number of events entering the storage system is managed according to the occurrence time of the events, and the storage system can be searched and updated accurately and effectively.

Abstract: Transmission of data for object storage, such as stream transmission for object storage, is disclosed. For instance, a group of objects acquired from an object layer is stored in a storage space for storing an object stream and serves as a first part of the object stream; and, in response to that an event related to at least one object in the group of objects occurs at a client terminal, the event-related information is stored in the storage space and is used as a second part of the object stream, the event-related information including at least one of the following: an identifier of the at least one object, a type of the event, and metadata of the event.

Abstract: Embodiments of the present disclosure relate to processing data. An example method includes acquiring data related to a first moment in streaming data of an object to be processed. The method further includes storing the data in a first entry of a data table based on an identification of the object to be processed, wherein the data table further includes a second entry before the first entry, and the second entry stores data related to a second moment before the first moment in the streaming data. The method further includes updating an index related to the object to be processed based on the first entry. Thus, a solution to the problem of performing search in data at different moments is provided, and it is unnecessary for a user to participate in the solution, thus improving the user experience and reducing the use of storage resources.

Abstract: Techniques for processing an access request and updating a storage system are provided. For instance, a method comprises: receiving an access request for an object associated with a storage system, the storage system including a plurality of physical nodes, each of the plurality of physical nodes including at least one set of virtual units, each set of virtual units including at least one virtual unit; determining, from a plurality of sets of virtual units included in the plurality of physical nodes of the storage system, a target set of virtual units associated with the object; and determining, from the target set of virtual units, a target virtual unit corresponding to the object. With the technical solution of the present disclosure, not only a set of virtual units on a physical node may be easily split and merged, but also huge computing resources that need to be allocated may be saved, so better user experience may be brought about at a lower cost.