Data storage arrangement

Patent metadata

Field	Value
Publication number	US-10235369-B2
Application number	US-201314437735-A
Country	US
Kind code	B2
Filing date	Sep 6, 2013
Priority date	Oct 24, 2012
Publication date	Mar 19, 2019
Grant date	Mar 19, 2019

Abstract

A computer arrangement includes a plurality of cluster systems each cluster system configured to archive data from at least one data processing installation, wherein each of the plurality of cluster systems is of modular design and includes at least one first component computer that receives data to be archived from the data processing installation, at least one mass memory system that buffer-stores the data to be archived, a second component computer that backs up the data to be archived on at least one further mass memory apparatus, and a cluster controller that controls the individual component computers of the respective cluster system; at least one data connection for data-oriented coupling of the plurality of cluster systems; and at least one composite controller that queries status data via a query interface of the cluster controllers of the plurality of cluster systems and transmits work orders to a control interface of the cluster controllers of the plurality of cluster systems.

First claim

The invention claimed is:

1. A computer arrangement having a multilevel, hierarchical architecture, the computer arrangement comprising:
a plurality of cluster systems including at least a first and a second cluster system on a first level of the multilevel, hierarchical architecture, each cluster system configured to archive data from at least one data processing installation, wherein each of the plurality of cluster systems is of modular design and comprises at least one first component computer, at least one mass storage system comprising a disk storage subsystem, a second component computer, and a cluster controller;
at least one data connection for data-oriented coupling of the plurality of cluster systems, comprising at least one control network that interchanges status queries and work orders between at least one composite controller and the cluster controllers of the plurality of cluster systems and a storage network that interchanges buffer-stored data associated with the interchanged work orders between the mass storage systems of the plurality of cluster systems; and
the at least one composite controller in the form of a software component executed by at least one component computer of at least one cluster system of the plurality of cluster systems on a second level of the multilevel, hierarchical architecture, the at least one composite controller and the cluster controller of the at least one cluster system forming a piece of a middleware of the at least one cluster system, the at least one composite controller configured for distribution of work orders between the plurality of cluster systems of the computer arrangement, wherein the work orders are instructions for an operation to back up and archive data; wherein
metadata and data to be archived are separated based on data type;
the first component computer of the first cluster system receives the data to be archived from the at least one data processing installation, the data to be archived belonging to a work order associated with the first cluster system;
the at least one mass storage system of the first cluster system buffer-stores the data to be archived;
the at least one composite controller queries status data of the first and second cluster system via the at least one control network and at least one query interface of the cluster controllers of the first and second cluster system;
upon occurrence of a predetermined system event, the at least one composite controller transmits a first order to stop execution of the work order associated with the first cluster system via the at least one control network to a control interface of the cluster controller of the first cluster system, transmits a second order to transmit the buffer-stored data associated with the stopped work order via the at least one control network to the control interface of the cluster controller of the first and second cluster system, and transmits a third order to execute the stopped work order via the at least one control network to the control interface of the cluster controller of the second cluster system; and wherein
upon receipt of the second order, the cluster controller of the first and second cluster system transmits data buffer-stored by the at least one mass storage system of the first cluster system to the second cluster system via the storage network; and
upon receipt of the third order, the cluster controller of the second cluster system controls the second component computer of the second cluster system to back up the data to be archived on at least one mass storage apparatus allocated to the second cluster system.

2. The computer arrangement according to claim 1 , in which each cluster system executes the middleware so that each cluster system can make queries to the query interfaces of other cluster systems of the computer arrangement and can transmit work orders to the control interfaces of other cluster systems of the computer arrangement.

3. The computer arrangement according to claim 1 , in which the at least one composite controller provides a user interface for operator control of the computer arrangement comprising the plurality of cluster systems, wherein the user interface compiles an overall state of the computer arrangement on the basis of results from status requests to the query interfaces of the plurality of cluster systems.

4. The computer arrangement according to claim 3 , in which the user interface further selects a cluster system to execute a work order, received via the user interface, on the basis of a utilization level of the individual cluster systems and of a predetermined rule database and transmits the work order to the control interface of the selected cluster system.

5. The computer arrangement according to claim 1 , in which the composite controller furthermore uses the at least one query interface of the cluster controllers to monitor an operating state of the individual cluster systems, to be prompted by identification of an operating fault in the first cluster system to transmit the first order to stop execution of a work order associated with the first cluster system to the control interface of the cluster controller of the first cluster system, transmit the second order to transmit buffer-stored data associated with the stopped work order from the first cluster system via the at least one data connection to the second cluster system to the control interface of the cluster controller of the first and second cluster system, and transmit the third order to execute the stopped work order to the control interface of the cluster controller of the second cluster system.

6. The computer arrangement according to claim 1 , in which the composite controller and the cluster controllers furthermore manage all of the mass storage systems that buffer-store the data to be archived from the plurality of cluster systems using a network file system jointly used by all cluster controllers.

7. The computer arrangement according to claim 1 , in which the cluster controller of the first cluster system transmits the data buffer-stored in the first cluster system to the second cluster system via the storage network when the predetermined event occurs.

8. The computer arrangement according to claim 1 , wherein the at least one mass storage apparatus comprises a tape drive.

Assignees

• Fujitsu Ltd

Inventors

• Werner Steffen

Classifications

• G06F11/1446

  Point-in-time backing up or restoration of persistent data · CPC title

• G06F17/30073 Primary

  Physics · mapped topic

• G06F3/0646

  Horizontal data movement in storage systems, i.e. moving data in between storage devices or systems · CPC title

• G06F2201/84

  Using snapshots, i.e. a logical point-in-time copy of the data · CPC title

• G06F2201/80

  Database-specific techniques · CPC title