Classifications

• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F11/00—Error detection; Error correction; Monitoring
  • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
  • G06F11/14—Error detection or correction of the data by redundancy in operation
  • G06F11/1402—Saving, restoring, recovering or retrying
  • G06F11/1446—Point-in-time backing up or restoration of persistent data
  • G06F11/1456—Hardware arrangements for backup
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F11/00—Error detection; Error correction; Monitoring
  • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
  • G06F11/14—Error detection or correction of the data by redundancy in operation
  • G06F11/1402—Saving, restoring, recovering or retrying
  • G06F11/1446—Point-in-time backing up or restoration of persistent data
  • G06F11/1458—Management of the backup or restore process
  • G06F11/1464—Management of the backup or restore process for networked environments

Definitions

• the present invention relates to the field of computer backup and in particular a system for automatically backing up and restoring data from a plurality of host devices in a heterogeneous environment.
• the simplest is a local architecture, intended for the direct and independent backup on each computer equipment of the data on one or more disks or local supports. This solution is of course not the most efficient because it requires the acquisition of a complete backup unit for each piece of equipment, and does not make it possible to optimize the use of the capacities of the backup media.
• client / server backup architectures There are also known “client / server” backup architectures.
• Patent WO99 / 66401 describing a data server for backing up and restoring data in a heterogeneous environment is also known in the prior art.
• the aim of the present invention is to propose a system for backing up disk data on a magnetic medium, allowing centralized backup and pooling of data with a view to optimizing the capacities of magnetic media.
• the invention relates to a system for automatically backing up and restoring data from a plurality of host devices in a heterogeneous environment, comprising a backup server comprising at least one hard disk buffer, the backup server being connected on the one hand to each of the computer equipment by at least one computer network and on the other hand to physical backup units constituted by disk management units and by tape management units, the backup server being controlled by a supervisor program ensuring the communication between the various elements of the system and the processing of the digital data exchanged, characterized in that the backup server comprises an emulation program for the simulation of backup equipment compatible with the environment of the '' any computer equipment to be backed up, and to record an image of a at least part of the disks of the host equipment and to transform the data exchanged with the different host equipment into data in a single format for exchanges with the backup server.
• the backup system makes it possible to automatically save, according to criteria such as the time, the date, subsets such as "volumes” (in the sense of Microsoft TM [registered trademark] of this term) or "disk partitions” (in the sense of Unix TM [registered trademark] of this term).
• the server comprises means for encapsulating the data originating from the host devices, for temporarily recording the data thus encapsulated on buffer disks, for grouping the data recorded on the buffer disks and transmitting them to the backup server.
• the server is connected to the external equipment by a first dedicated high-speed link for the exchange of data to be saved or restored, and by a second link for the exchange of control signals.
• the invention also relates to a method for automatically backing up and restoring data from a plurality of host devices in a heterogeneous environment, comprising a step of activation by a server of the transmission of the data to be saved recorded on the memory of a equipment external to a server for saving and recording said data to be saved on a magnetic medium, characterized in that it includes a step of replicating on a memory of the backup server the physical medium of the memory of each of the host equipments, and a step of processing the data thus replicated for recording on at least one magnetic medium associated with the backup server.
• the replication step consists of saving a copy of the volumes or volumes on a hard disk of the backup server. disk partitions of at least part of the host equipment.
• the transmission of the data to be saved between the external equipment and the temporary disk is achieved by a high speed link, such as the fiber optic bus (Fiber Channel).
• a high speed link such as the fiber optic bus (Fiber Channel).
• the replication step consists in temporarily suspending the “mirror” function of the RAID 1 (registered trademark) procedure of the host equipment and in recording on a temporary memory the data of the separate part of the mirror of the disk of the host equipment, and rebuilding the mirror after the completion of data replication.
• RAID 1 registered trademark
• FIG. 1 shows the block diagram of a system according to the invention.
• the architecture of an automatic data backup and restoration system comprises:
• control unit (1) managing the user interface and the database describing the collection of magnetic backup media.
• host disk management units (2, 3) which transfer the data from the host disks (4 to 7) to the bus (8) of the backup server and vice versa.
• the control unit (1) controls the host disk management units (2, 3) and the tape management units (9).
• a disk management system User data from the hard disks of the host devices (4 to 7) is accessed through a disk management system.
• This management system has mirror (RAID 1 [registered trademark]) and temporary mirror suspension functions, allowing the connection of a host on the separate part of the mirror.
• the management of modifications (bit-map) during suspension is necessary to allow the mirror to be quickly brought back into conformity.
• User data is managed with this system. The user continues to work on a branch of the mirror. The other contains the data to be saved. Once the data has been saved, the mirror is reconstituted and the user finds the initial situation.
• the communications between the various elements of the backup server are provided by a fiber optic bus (8) for networks extending over a few hundred or thousands of meters, or copper for networks of small geographic extension (a few tens of meters). Bus speeds are typically around 100 megabytes in [half duplex]. In case higher flow rates are required, the system can integrate several buses. It is also possible a single machine implementation, therefore without bus.
• the control unit (1) controls the various elements of the system, through a local network. She understands :
• - a database containing the links between the backup name, the date, the volumes and magnetic media as well as the location of each magnetic media in the robots.
• the database also contains the definition of each backup occurrence, defined by its name, and its criteria for implementation (date, time, circumstances) and conservation (date and / or expiration conditions). This database may possibly be split for security reasons.
• - means of remote control of the host machines.
• the backup server needs to know the configuration of user disks, which is scalable. When a backup is requested, we build the topology (location of the physical disks, start and end address of the different volumes from the user request which is formed by a sequence of volume names. For this, client software installed in each of the host devices transmits to the backup server the description of the structure of each disk in response to a request from the backup server.
• control unit From the name of the backup, the control unit establishes a backup plan, containing the list of host disk management units, disk units and disk addresses involved in the backup. The same operation is performed for restoring the data. The data does not pass through the control unit.
• - bus management means These means manage a permanent dialogue between the host disk units, the tape management units and the status of the current data transfers (active, recovery, waiting), the bit rates, the causes of the waits, etc.
• means of remote maintenance and remote diagnosis comprising a local network or an asynchronous line associated with a modem.
• the host disk management unit allows the backup server to connect to a configuration disk management unit and transfer the data to be backed up.
• the connections to the disks are made through:
• Fiber Channel a specific high speed fiber optic bus (Fiber Channel).
• the host disk management unit controls the transfer of data to the backup device using the tape management unit.
• Host disk management units. are seen by disk management systems as a host in the sense of the operating system of the host equipment.
• the host disk management unit is responsible for processing data packaging (header and end of block) and sending this data to the management unit of the magnetic media control robot. The data must be kept for the time necessary for them to be taken into account by the robot management unit.
• the tape management unit manages the backup devices (tape drive [streamers]) and the robot or robots containing them.
• the tape management unit allows the data to be backed up to be transferred at a maximum speed, taking into account the characteristics of the disks and the bus type of the configuration. Connections to backup devices are made using [SCSI] buses and ESCON (registered trademark), high speed fiber optic (Fiber Channel) channels.
• the tape management unit receives orders from the control unit allowing robot management: selection of a support, change of support, ...
• the tape management unit receives commands and data from the host disk management unit. Acknowledgment of receipt of data is sent to the transmitter. During a restore, the tape management unit controls the transfer of data to the host disks using the host disk management unit.

Landscapes

• Engineering & Computer Science (AREA)
• Theoretical Computer Science (AREA)
• Quality & Reliability (AREA)
• Physics & Mathematics (AREA)
• General Engineering & Computer Science (AREA)
• General Physics & Mathematics (AREA)
• Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
• Computer And Data Communications (AREA)
• Debugging And Monitoring (AREA)

Applications Claiming Priority (3)

Publications (1)

Family

ID=8849756

Family Applications (1)

Country Status (5)

Families Citing this family (8)

Family Cites Families (6)

• 2000
  • 2000-04-28 FR FR0005527A patent/FR2808349A1/fr active Pending
• 2001
  • 2001-04-27 JP JP2001580623A patent/JP2003532219A/ja active Pending
  • 2001-04-27 WO PCT/FR2001/001324 patent/WO2001084265A2/fr not_active Application Discontinuation
  • 2001-04-27 EP EP01929756A patent/EP1285320A2/de not_active Ceased
  • 2001-04-27 AU AU2001256442A patent/AU2001256442A1/en not_active Abandoned

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events