Deployment of a software image on multiple targets with streaming technique

    公开(公告)号:GB2512006A

    公开(公告)日:2014-09-17

    申请号:GB201412336

    申请日:2012-12-04

    Applicant: IBM

    Abstract: A solution is proposed for deploying a software image from a source data- processing system on a plurality of target data-processing entities of a target data- processing system, the software image comprising a plurality of memory blocks being individually accessible, with a predefined subset of the memory blocks defining a bootstrap module. A corresponding method (400) comprises the steps of downloading (406) the bootstrap module onto a main one of the target data- processing entities from the source data-processing system, booting (407-409) the main target data-processing entity from the bootstrap module thereby loading a streaming driver comprised in the bootstrap module, serving (410-422) each request of accessing a selected memory block of the software image on the main data- processing entity by the streaming driver, the streaming driver downloading (413- 414) the selected memory block from the source data-processing system and storing (415-416) the selected memory block into the main target data-processing entity in response to a first one of the requests of accessing the selected memory block, or accessing (417) the selected memory block in the main target data-processing entity otherwise, providing (423-427) the bootstrap module to each one of a set of secondary ones of the target data-processing entities, booting (428-429) each secondary target data-processing entity from the bootstrap module thereby loading the streaming driver, and serving (430-448) each request of accessing a further selected memory block of the software image on the secondary data-processing entity by the streaming driver, the streaming driver retrieving (433-440) the further selected memory block from the main target data-processing entity and storing (441- 442) the further selected memory block into the secondary target data-processing entity in response to a first one of the requests of accessing the further selected memory block, or accessing (443) the further selected memory block in the secondary target data-processing entity otherwise.

    Meta-systems management tool
    12.
    发明专利

    公开(公告)号:GB2502539A

    公开(公告)日:2013-12-04

    申请号:GB201209547

    申请日:2012-05-30

    Applicant: IBM

    Abstract: A method for managing computing components comprising: collecting management information about the computing components, such as applications or hardware components, using a data collection component, for example reading databases associated with each component storing objects of the collected management information in a full-text search-index data-structure, generating, for each collected object, a displayable result snippet, searching the full-text search-index data-structure for objects which match a received parameter, retrieving all matching objects of the full-text search-index data-structure related to the search, and displaying the matching objects in form of the related result snippets.

    Modifying the order of checking virtual machines for cached disc data

    公开(公告)号:GB2502288A

    公开(公告)日:2013-11-27

    申请号:GB201208959

    申请日:2012-05-22

    Applicant: IBM

    Abstract: A number of virtual machines 106 are connected to a storage server 118 over a network 116. Each virtual machine caches data from the server. Each virtual machine has a list of other virtual machines. When a virtual machine attempts to read data from the server, it checks each virtual machine on its list to see if it has the data in its cache. The virtual machines are checked in an order given by the ranking in the list. If the data is in the cache of a virtual machine, it is copied from that cache rather than being read from the server. In this case, the ranking of the server with the data is updated. If the data is not found in any of the virtual machine caches, then the data is read from the server.

Patent Agency Ranking