System, a Method and a Device for Updating a Data Set Through a Communication Network

• US 20070255766A1
• Filed: 08/12/2004
• Published: 11/01/2007
• Est. Priority Date: 08/12/2004
• Status: Active Grant

• Current Assignee: Telecom Italia S.p.A.
• Original Assignee: Telecom Italia S.p.A.
• Inventors: Girardi, Raffaele, Maffione, Eugenio Maria, Di Giglio, Andrea
• Granted Patent: US 7,987,154 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/2074 Asynchronous techniques
• CPC Class Codes: G06F 11/2076 Synchronous techniques

Abstract

A system for updating a data set through a geographical communication network. The data set is stored in a recovery storage unit located in a recovery data centre and has to be updated with data blocks generated by a processing system located in a primary data centre. The recovery data centre is associated through the geographical communication network to a device provided with a temporary storage area for temporarily storing in corresponding storage locations the incoming data blocks generated by the processing system and directed to update the data set, wherein a first synchronous acknowledgement signal is provided to the primary data centre acknowledging the successful writing in the temporary storage area of the data blocks as soon as the data blocks are written in the temporary storage area so as not to block or slow down the normal way of operating of the processing system, and wherein the locations of the temporary storage area occupied by the data blocks are made available for new incoming data blocks when a second acknowledgement signal is received acknowledging the successful updating of the data set.

Claims

1-34. -34. (canceled)

35. A system for updating a data set through a communication network, said data set being stored in a recovery storage unit located in a recovery data centre and having to be updated with incoming data blocks generated by at least one processing system associated with a primary data centre, comprising:

a device interposed between said primary data centre and said recovery data centre through said communication network and associated with a temporary storage area for temporarily writing in corresponding storage locations the incoming data blocks generated by said processing system and for forwarding said data blocks to said recovery data centre wherein said device comprises modules capable of;

sending a first synchronous acknowledgement signal to said primary data centre acknowledging the successful writing in said temporary storage area of said data blocks as soon as said data blocks are written in said temporary storage area,

receiving and managing a second acknowledgement signal sent by said recovery data centre acknowledging the successful updating of said data set following the forwarding of said data blocks, and

making available for new incoming data blocks the locations of said temporary storage area following the receipt of said second acknowledgement signal.

36. The system according to claim 35, wherein said processing system is associated with a primary storage unit where a copy of said data set is stored.

37. The system according to claim 36, wherein said temporary storage area has a storage capability lower than the storage capability both of said primary storage unit and of said recovery storage unit.

38. The system according to claim 35, wherein the distance between said device and said primary data centre is lower than the distance between said device and said recovery data centre.

39. The system according to claim 35, wherein said device further comprises modules capable of writing said data blocks in said storage area into an address requested by said primary data centre as follows:

if the data blocks at the requested address are not written in the storage area, writing said data blocks in said storage area, if the data blocks at the requested address are already written in the storage area, but said data blocks have not been forwarded, writing said data blocks in said storage area by replacing the previous data blocks; and

if the data blocks at the requested address are already written in the storage area have been forwarded and the acknowledgement from said recovery data centre has not been received, writing said data blocks in a new location in said storage area.

40. The system according to claim 35, wherein said communication network comprises at least one protocol selected from the group of an iSCSI type protocol and a fibre channel type protocol.

41. The system as claimed in claim 35, wherein said device comprises:

interface modules for interfacing said device to said primary data centre and to said recovery data centre respectively;

a filter module for intercepting I/O commands sent by said primary data centre and for selectively sending said commands either to said temporary storage area for temporarily storing said data blocks or to said recovery data centre; and

a control module for receiving said data blocks with their relative address parameters from said primary data centre, encapsulating said data blocks inside commands reserved for the management of said recovery storage unit and forwarding them to said recovery data centre.

42. The system according to claim 35, wherein said device comprises:

a network interface card;

an iSCSI target device arranged downstream of said network interface card;

a modified SCSI target device arranged downstream of said iSCSI target device;

a cache system associated with said modified iSCSI target device;

a logical volume manager or LVM device arranged downstream of said modified SCSI target device;

a native SCSI subsystem arranged downstream of said logical volume manager device;

an iSCSI initiator device arranged downstream of said SCSI subsystem; and

a further network interface card toward said recovery data centre.

43. The system according to claim 41, wherein said device comprises:

a host bus adapter toward said primary data centre;

a modified SCSI target device, arranged downstream of said host bus adapter;

a cache system associated with said modified SCSI target device;

a logical volume manager or LVM device arranged downstream of said modified SCSI target device;

an SCSI subsystem arranged downstream of said logical volume manager device; and

a further host bus adapter toward said recovery storage unit.

44. The system according to claim 35, wherein said device for temporarily storing said data blocks comprises a network device associated with said communication network.

45. The system as claimed in claim 35, wherein storage units are disk storage units.

46. The system as claimed in claim 35, wherein said temporary storage area comprises a plurality of storage areas respectively dedicated to a plurality of processing systems and/or to a plurality of primary data centres.

65. A communication network associated with a system for updating a data set as claimed in claim 35.

47. A method for updating a data set through a communication network, said data set being stored in a recovery storage unit located in a recovery data centre and having to be updated with incoming data blocks generated by at least one processing system associated with a primary data centre, comprising the steps of:

temporarily writing through said communication network the incoming data blocks generated by said processing system in corresponding storage locations of a temporary storage area associated with a device interposed between said primary data centre and said recovery data centre;

providing by said device and through said communication network a first synchronous acknowledgement signal to said primary data centre acknowledging the successful writing in said temporary storage area of said data blocks as soon as said data blocks are written in said storage area;

forwarding by said device and through said communication network said data blocks to said recovery data centre;

receiving and managing by said device and through said communication network a second acknowledgement signal acknowledging the successful updating of said data set into said recovery data centre; and

making available for new incoming data blocks the locations of said temporary storage area following the receipt of said second acknowledgement signal.

48. The method according to claim 47, comprising the step of:

storing said data set into a primary storage unit associated with said processing system.

49. The method according to claim 47, wherein the step of temporarily writing the incoming data blocks in said device comprises the steps of:

writing said data blocks in said storage area if the data blocks at the requested address are not written in the storage area;

writing said data blocks in said storage area by replacing the previous data blocks, if the data blocks at the requested address are already written in the storage area, but said data blocks have not been forwarded; and

writing said data blocks in a new location in said storage area if the data blocks at the requested address are already written in the storage area have been forwarded and the acknowledgement from said recovery data centre has not been received.

50. The method according to claim 49 wherein the step of temporarily writing the incoming data blocks in said storage area comprises the further step of:

writing by replacing the data blocks which have not been forwarded yet, if the data blocks at the requested address are stored twice in said storage area.

51. The method according to claim 47, wherein said step of forwarding said data blocks from said device to said recovery data centre comprises the steps of:

polling said storage area associated with said device for checking whether there are any data blocks to be forwarded to said recovery data centre;

sending to said recovery data centre the data blocks to be forwarded and simultaneously starting a timer having a predetermined timeout delay; and

wherein the step of receiving and managing said second acknowledgement signal comprises the steps of;

checking whether said timeout delay has elapsed;

incrementing a counter representative of lost data blocks if said timeout has elapsed; and

sending an alert to the primary data centre for locking any further transmission of data blocks to said device if said counter of lost data blocks has reached a predetermined value.

52. The method according to claim 51, wherein the step of receiving and managing said second acknowledgement signal comprises the step of managing a negative acknowledgement signal indicating that said data blocks have arrived to the recovery data centre, but are corrupted.

53. The method according to claim 51, wherein said timeout delay is set to a value greater than the round trip time for a data block to be sent from said device and received by said recovery storage unit.

54. The method according to claim 47, for recovering data blocks toward the primary data center, further comprising the steps of:

requesting recovered data blocks by said device to said recovery data centre, if the data blocks at the requested address are not stored in said temporary storage area;

reading recovered data blocks from the temporary storage area if the data blocks at the requested address are stored in said storage area and they have not undergone the forwarding step to the recovery data centre;

reading recovered data blocks from the temporary storage area if the data blocks at the requested address are stored in said temporary storage area and they have been forwarded to the recovery data centre; and

recovering said recovered data blocks by said device toward said primary data centre through at least one of said previous steps on request of said primary data centre.

66. A computer program product loadable into the memory of at least one electronic computer and comprising portions of software code capable of implementing the method as claimed in claim 47.

55. A device for updating a data set through a communication network, said device being associated with a storage area for temporarily writing in corresponding storage locations incoming data blocks generated by a processing system associated with a primary data centre and for forwarding said data blocks to a recovery storage unit located in a recovery data centre, comprising modules capable of:

sending through said communication network a first synchronous acknowledgement signal to said primary data centre acknowledging the successful writing in said temporary storage area of said data blocks as soon as said data blocks are written in said temporary storage area,

receiving through said communication network and managing a second acknowledgement signal sent by said recovery data centre acknowledging the successful updating of said data set following the forwarding of said data blocks, and

making available for new incoming data blocks the locations of said temporary storage area following the receipt of said second acknowledgement signal.

56. The device according to claim 55, wherein said processing system is associated with a primary storage unit where a copy of said data set is stored.

57. The device according to claim 55, wherein the distance between said device and said primary data centre is lower than the distance between said device and said recovery data centre.

58. The device according to claim 55, comprising modules capable of interfacing at least one protocol selected from the group of an SCSI type protocol, an iSCSI type protocol and a fibre channel type protocol.

60. The device according to claim 55, comprising:

a network interface card toward said primary data centre;

an iSCSI target device, arranged downstream of said network interface card or host bus adapter;

a modified SCSI target device arranged downstream of said iSCSI target device;

a cache system associated with said modified SCSI target device;

a logical volume manager or LVM device arranged downstream of said modified SCSI target device;

a native SCSI subsystem arranged downstream of said logical volume manager device;

an iSCSI initiator device arranged downstream of said SCSI subsystem;

a further network interface card or host bus adapter toward said recovery data centre.

61. The device according to claim 55, comprising:

a host bus adapter toward said primary data centre;

a modified SCSI target device arranged downstream of said host bus adapter;

a cache system connected to said modified SCSI target device;

a logical volume manager or LVM device, arranged downstream of said modified SCSI target device;

an SCSI subsystem arranged downstream of said logical volume manager device; and

a further host bus adapter toward said recovery data centre.

62. The device according to claim 60, wherein said LVM device is replaced by a software virtualisation device.

63. The device according to claim 61, wherein said LVM device is replaced by a software virtualisation device.

64. The device according to claim 55, comprising a network device associated with said communication network.

59. A device according 55, comprising:

interface modules for interfacing said device to said primary data centre and to said recovery data centre respectively;

a filter module for intercepting an I/O command to which data are associated and for deciding whether to send said command either to a storage area for temporarily storing said blocks or to said recovery storage unit; and

a control module for receiving said data blocks with their relative address parameters, encapsulating them inside commands reserved for the management of said recovery storage unit and sending them to said recovery data centre.

67. A method of providing a synchronous mirroring service for updating a data set through a communication network, said data set being stored in a recovery storage unit located in a recovery data centre and having to be updated with data blocks generated by at least one client system associated with a primary data centre, comprising the steps of:

temporarily writing through said communication network incoming data blocks generated by said at least one client system in corresponding storage locations of a temporary storage area associated with a device interposed between said primary data centre and said recovery data centre;

providing by said device and through said communication network a first synchronous acknowledgement signal to said primary data centre acknowledging the successful writing in said temporary storage area of said data blocks as soon as said data blocks are written in said storage area;

forwarding by said device and through said communication network said data blocks to said recovery data centre;

receiving and managing by said device and through said communication network a second acknowledgement signal acknowledging the successful updating of said data set into said recovery data centre; and

making available for new incoming data blocks the locations of said temporary storage area following the receipt of said second acknowledgement signal.

68. The method of providing a synchronous mirroring service according to claim 67, further comprising the steps of:

connecting a plurality of primary data centres to said device for temporarily writing respective data blocks generated by respective client systems associated with said plurality of primary data centres, said device being arranged in a metropolitan area type network;

connecting said metropolitan area type network to a wide area type network; and

forwarding said respective data blocks from said device through said wide area type network to said recovery data centre.

69. The method of providing a synchronous mirroring service according to claim 68, comprising the further step of:

assessing priority in the process of transferring said respective data blocks from said device through said wide area type network to said recovery data centre with respect to respective data blocks of other client systems in case of a client system failure.