Distributed namespace aggregation

• US 20070198458A1
• Filed: 02/06/2006
• Published: 08/23/2007
• Est. Priority Date: 02/06/2006
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Pudipeddi, Ravisankar
• Granted Patent: US 7,640,247 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/10 File systems; File servers
• CPC Class Codes: Y10S 707/99952 Coherency, e.g. same view to multiple users

Abstract

Aspects of the subject matter described herein relate to distributed namespace aggregation. In aspects, a distributed file system is extended to allow multiple servers to seamlessly host files associated with aggregated links and/or aggregated roots. A request for a directory listing of an aggregated link or root may cause a server to sniff multiple other servers that host files associated with the link or root to create and return a concatenated result. Sniffing may also be used to determine which servers host the file to which the client is requesting access. Altitude may be used to determine which servers to make visible to the client and may also be used to determine which servers are in the same replica group and which are not.

Claims

1. A computer-readable medium having computer-executable instructions, comprising:

receiving a request to open a file of a distributed file system;

determining that the file is associated with an aggregated link, wherein the aggregated link comprises a path that is associated with at least two folders that are not replicas of each other; and

determining whether to generate a referral.

2. The computer-readable medium of claim 1, wherein the request includes a complete universal naming convention (UNC) path to the file.

3. The computer-readable medium of claim 1, further comprising:

receiving a request for the referral;

constructing the referral to include paths on servers at which the file is accessible, wherein the paths include a server from each replica group associated with the aggregated link.

4. The computer-readable medium of claim 3, wherein the paths are ordered to load balance requests to the servers.

5. The computer-readable medium of claim 4, wherein the file comprises a directory.

6. The computer-readable medium of claim 5, further comprising

receiving a request to query directory information associated with the file;

requesting directory information from a server from each replica group associated with the aggregated link;

concatenating the directory information into a response; and

sending the response.

7. The computer-readable medium of claim 6, wherein concatenating the directory information into a response comprises discarding conflicting directory information.

8. The computer-readable medium of claim 7, wherein discarding conflicting directory information comprises determining altitudes of the servers associated with the directory conflicting directory information, wherein one server has a highest altitude, and discarding directory information from servers with altitudes less than the highest altitude.

9. A method implemented at least in part by a machine, comprising:

receiving a request to open a file of a distributed file system;

determining that the file is associated with an aggregated root path, wherein the aggregated root path is associated with at least two root folders that are not replicas of each other; and

determining whether to generate a referral.

10. The method of claim 9, wherein the file is represented by a universal naming convention (UNC) path, and further comprising sniffing the at least two root folders to find a component of the UNC path that identifies at least one of the folders.

11. The method of claim 10, further comprising providing a referral that includes paths of root directories on servers that host the at least one of the folders.

12. The method of claim 11, further comprising ordering the paths according to network cost.

13. The method of claim 9, further comprising when more than one of the at least two root folders include an identical beginning portion of a path, determining which of the at least two folders to reference in the referral based on altitudes assigned to each of the folders.

14. The method of claim 13, wherein each altitude comprises a number, wherein folders assigned to equal altitudes are part of a single replica group.

15. The method of claim 9, further comprising aggregating directories from multiple servers that host the at least two root folders into a list and returning the list.

16. An apparatus for servicing requests in a distributed file system, comprising:

a communication component arranged to communicate according to a distributed file system protocol;

a server component arranged to examine operations sent from the communication component to a local file system; and

a service arranged to provide information to the server component for aggregation-related requests.

17. The apparatus of claim 16, wherein the server component comprises a file system filter arranged to execute in kernel mode.

18. The apparatus of claim 16, wherein the service comprises a process arranged to execute in user mode.

19. The apparatus of claim 16, further comprising a database arranged to store results from sniffing on other servers of the distributed file system.

20. The apparatus of claim 16, wherein the service is arranged to aggregate a directory structure of an entire distributed file system into a single namespace.