Loading patent…
AssigneeEMC Corporation
InventorsMichael Paul Wagner, Paul Thomas McGrath, J. Brandon Myers, Bruce Allen Keesee
Estimating the term…
A method allows point-in-time viewing of session data for multiple sessions being stored on a disk. Session bits are set for the session data in memory which points to the data at any particular point in time, and this can be done for multiple sessions. In order to make the session data persistent, separate session bits are stored in a portion of a disk reserved for such session bits. Thus, in the event of failure where session bits which are stored in volatile memory such as RAM, are lost, the duplicate copy set in the non-volatile memory can be used to recreate the session once the failure is corrected.
A method of allowing point-in-time view of data on a disk, comprising:
A method of allowing point-in-time views of data on a disk, comprising:
A method of allowing point-in-time view of data on a disk, comprising:
Dependencies are inferred from the claim text.
From the US filing history. Foreign family members aren’t listed.
Backward citations are indexed for grants only.
Drawings are temporarily unavailable. Open the PDF at USPTO
At system level: resetting or repowering
At system level: restarting or rejuvenating
Coherency, e.g. same view to multiple users: recoverability
Coherency, e.g. same view to multiple users: archiving or backup
Term end not estimated
This invention relates to a method of creating a point-in-time copy of data on a disk, i.e., a Logical Unit, which is accessible from a different host. More specifically, the invention allows views of multiple point-in-time copies in a manner in which access to such point-in-time copies is maintained notwithstanding multiple failures such as failure of a processor, a reboot operation, power failure, etc.
In current storage environments or systems, such as block storage systems, it is often the case that such environments or systems will include one or more hosts, such as a server, for example, such as is available from Sun Microsystems and/or other companies. Such hosts are typically attached through network connections, such as a fibre channel, to one or more storage units or arrays. Such storage units or arrays are typically configured as a cabinet including multiple storage disk drives, and onboard intelligence in the form of one or more processors which operate to control the storage unit. Examples of such storage units are those available from EMC Corporation, for example, under the product name Clariion™.
In such environments, multiple input-output (I/O) operations occur due to users working through one or more hosts to read and write data to disks in the storage unit.
Having thus briefly described the invention, the same will become better understood from the appended drawings, wherein:
FIG. 1 is a block diagram illustrating a host attached through a fibre channel switch to a storage array, having multiple disks contained therein;
FIG. 2 is a block diagram illustrating a portion of FIG. 1 as a phantom Logical Unit in which session bits are used to determine which data to present;
To facilitate understanding of this invention, the following terms are set forth as defined terms, which will then facilitate an understanding of the method described herein.
Chunk Map Entry: This refers to a structure used to indicate what chunks have had COFW operations performed. These entries are stored in a hash table for fast lookup. Their existence indicates that a COFW operation has occurred and there is data in the View Cache.
Chunk Size: This is the unit of measure used to improve performance. Instead of storing individual blocks in the View Cache and keeping internal management information about each individual block, an operating system driver will use multiple blocks (Chunk Size). All COFW (defined below) I/O is done in Chunk Sized I/Os. The default Chunk Size us 64 KB and is user configurable in selectable sizes of 16 KB, 32 KB, 64 KB, 128 KB, and 256 KB.