Fast path for performing data operations

• US 7,173,929 B1
• Filed: 08/13/2002
• Issued: 02/06/2007
• Est. Priority Date: 12/10/2001
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: Incipient Inc.
• Inventors: Testardi, Richard
• Primary Examiner(s): Vital; Pierre
• Assistant Examiner(s): Chery; Mardochee
• Application Number:
  US10/218,098
• Time in Patent Office: 1,638 Days
• Field of Search: None
• US Class Current: 370/355
• CPC Class Codes: G06F 3/0611 in relation to response time
• CPC Class Codes: G06F 3/0635 by changing the path, e.g. traffic rerouting, path reconfiguration
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]

Abstract

Described are techniques used in a computer system for handling data operations to storage devices. A switching fabric includes one or more fast paths for handling lightweight, common data operations and at least one control path for handling other data operations. A control path manages one or more fast paths. The fast path and the control path are utilized in mapping virtual to physical addresses using mapping tables. The mapping tables include an extent table of one or more entries corresponding to varying address ranges. The size of an extent may be changed dynamically in accordance with a corresponding state change of physical storage. The fast path may cache only portions of the extent table as needed in accordance with a caching technique. The fast path may cache a subset of the extent table stored within the control path. A set of primitives may be used in performing data operations. A locking mechanism is described for controlling access to data shared by the control paths.

Claims

1. A method executed in a computer system for processing a data operation comprising:

determining if said data operation has at least one predetermined criteria characterizing said data operation as being a commonly performed non-complex data operation using a primitive operation; and

routing said data operation to a fast path for processing if said data operation has said at least one predetermined criteria, and routine said data operation to a general control path for processing otherwise, wherein said fast path and said general control path are included in a switching fabric.

2. The method of claim 1, wherein said fast path and said general control path are implemented in at least one of:

hardware or software.

3. The method of claim 1, wherein said predetermined criteria is one of:

said data operation is within a single logical volume, said data operation involves data from within virtual address boundaries of a single mapping table entry, said data operation is a read operation, and said data operation is a write operation.

4. The method of claim 1, wherein said predetermined criteria is one of:

said data operation is indicated as allowable in accordance with information included in a mapping table used in mapping virtual to physical locations for said data operation, and said data operation does not cause a serialization conflict in connection with a mirrored device.

5. The method of claim 1, further comprising:

routing said data operation to said general control path by causing a fault if said data operation does not have said at least one predetermined criteria.

6. The method of claim 1, further comprising:

routing said data operation to said general control path in accordance with information included in a mapping table associated with a virtual address to physical address translation of an address of said data operation.

7. The method of claim 6, further comprising:

determining a volume segment descriptor associated with said data operation, said volume segment descriptor including said mapping table.

8. The method of claim 1, wherein said switching fabric includes a plurality of general control paths, each of said general control paths associated with at least one fast path.

9. The method of claim 8, further comprising:

each of said plurality of general control paths controlling said at least one associated fast path.

10. The method of claim 9, wherein said switching fabric includes a plurality of virtualization engines, each of said plurality of virtualization engines including at least one general control path and at least one associated fast path.

11. The method of claim 10, wherein, for each virtualization engine that includes a plurality of general control paths, only one of said general control paths being active at a time with another of said plurality of general control paths being active if said only one general control path is off-line.

12. The method of claim 11, further comprising:

issuing said data operation from a host system to a virtualization engine which performs virtual to physical address translation of an address associated with said data operation.

13. The method of claim 12, wherein said data operation from said host is routed through a first switch that is directly connected to a fast path included in a second switch.

14. The method of claim 13, wherein said first switch is directly connected to said second switch by plugging into a port of said second switch.

15. The method of claim 1, wherein said data operation is issued by said fast path if said data operation has said at least one predetermined criteria and is otherwise issued by said general control path.

16. The method of claim 1, wherein said data operation is an upstream data operation from a host and the method further comprising:

producing at least one downstream data operation to a physical device associated with said data operation corresponding to said upstream data operation; and

creating an element on a pending data operation list corresponding to said data operation.

17. The method of claim 1, further comprising:

routing return data operation data using said fast path to a host that issued said data operation if there is no error, and otherwise routing return data operation data using said general control path to a host that issued said data operation.

18. A computer program product, tangibly embodied in a computer-readable medium, for processing a data operation, the computer program product including instructions being operable to cause a data processing apparatus to:

determine if said data operation has at least one predetermined criteria characterizing said data operation as being a commonly performed non-complex data operation using a primitive operation; and

route said data operation to a fast path for processing if said data operation has said at least one predetermined criteria, and routing said data operation to a general control path for processing otherwise, wherein said fast path and said general control path are included in a switching fabric.

19. The computer program product of claim 18, wherein said fast path and said general control path are implemented in at least one of:

hardware or software.

20. The computer program product of claim 18, wherein said predetermined criteria is one of:

said data operation is within a single logical volume, said data operation involves data from within virtual address boundaries of a single mapping table entry, said data operation is a read operation, and said data operation is a write operation.

21. The computer program product of claim 18, wherein said predetermined criteria is one of:

said data operation is indicated as allowable in accordance with information included in a mapping table used in mapping virtual to physical locations for said data operation, and said data operation does not cause a serialization conflict in connection with a mirrored device.

22. The computer program product of claim 18, further operable to cause the data processing apparatus to:

said data operation to said general control path by causing a fault if said data operation does not have said at least one predetermined criteria.

23. The computer program product of claim 18, further operable to cause the data processing apparatus to:

said data operation to said general control path in accordance with information included in a mapping table associated with a virtual address to physical address translation of an address of said data operation.

24. The computer program product of claim 21, further operable to cause the data processing apparatus to:

a volume segment descriptor associated with said data operation, said volume segment descriptor including said mapping table.

25. The computer program product of claim 18, wherein said switching fabric includes a plurality of general control paths, each of said general control paths associated 20 with at least one fast path.

26. The computer program product of claim 25, further wherein:

each of said plurality of general control paths comprises a computer program product tangibly embodied in a computer-readable medium for controlling said at least one associated fast path.

27. The computer program product of claim 26, wherein said switching fabric includes a plurality of virtualization engines, each of said plurality of virtualization engines including at least one general control path and at least one associated fast path.

28. The computer program product of claim 27, wherein, for each virtualization engine that includes a plurality of general control paths, only one of said general control paths being active at a time with another of said plurality of general control paths being active if said only one general control path is off-line.

29. The computer program product of claim 28, further operable to cause the data processing apparatus to:

said data operation from a host system to a virtualization engine which performs virtual to physical address translation of an address associated with said data operation.

30. The computer program product of claim 29, wherein said data operation from said host is routed through a first switch that is directly connected to a fast path included in a second switch.

31. The computer program product of claim 30, wherein said first switch is directly connected to said second switch by plugging into a port of said second switch.

32. The computer program product of claim 18, wherein said data operation is issued by said fast path if said data operation has said at least one predetermined criteria and is otherwise issued by said general control path.

33. The computer program product of claim 18, wherein said data operation is an upstream data operation from a host and the computer program product further operable to cause the data processing apparatus to:

at least one downstream data operation to a physical device associated with said data operation corresponding to said upstream data operation; and

an element on a pending data operation list corresponding to said data operation.

34. The computer program product of claim 18, further operable to cause the data processing apparatus to:

return data operation data using said fast path to a host that issued said data operation if there is no error, and otherwise route return data operation data using said general control path to a host that issued said data operation.