초록 ▼

A is a method is disclosed for integrating a file system with a RAID array that exports precise information about the arrangement of data blocks in the RAID subsystem. The file system examines this information and uses it to optimize the location of blocks as they are written to the RAID system. Thu

A is a method is disclosed for integrating a file system with a RAID array that exports precise information about the arrangement of data blocks in the RAID subsystem. The file system examines this information and uses it to optimize the location of blocks as they are written to the RAID system. Thus, the system uses explicit knowledge of the underlying RAID disk layout to schedule disk allocation. The method uses separate current-write location (CWL) pointers for each disk in the disk array where the pointers simply advance through the disks as writes occur. The algorithm used has two primary goals. The first goal is to keep the CWL pointers as close together as possible, thereby improving RAID efficiency by writing to multiple blocks in the stripe simultaneously. The second goal is to allocate adjacent blocks in a file on the same disk, thereby improving read back performance. The method satisfies the first goal by always writing on the disk with the lowest CWL pointer. For the second goal, a new disks chosen only when the algorithm starts allocating space for a new file, or when it has allocated N blocks on the same disk for a single file. A sufficient number of blocks is defined as all the buffers in a chunk of N sequential buffers in a file. The result is that CWL pointers are never more than N blocks apart on different disks, and large files have N consecutive blocks on the same disk.

대표청구항 ▼

1. A system, comprising: at least one processor; memory that stores instructions and data; and a RAID array; wherein said processor executes said instructions to allocate files in a file system for said RAID array, said instructions including the steps of: (a) selecting an inode for said

1. A system, comprising: at least one processor; memory that stores instructions and data; and a RAID array; wherein said processor executes said instructions to allocate files in a file system for said RAID array, said instructions including the steps of: (a) selecting an inode for said RAID array having at least one dirty block from a list of inodes having dirty blocks; (b) write allocating blocks of data referenced by said inode to a storage device in said RAID array wherein said storage device comprises a plurality of storage blocks and adding said blocks of data to a list of writeable blocks of data for said storage device; (c) determining if all nodes have been processed, and when all of said inodes in said list of inodes have not been processed, continue repeating steps a-b; and, (d) flushing all unwritten stripes to said RAID array; wherein said blocks of data referenced by said inode are write allocated in a depth-first, post visit traversal of said blocks of data. 2. The system of claim 1 wherein said blocks of data are stored in a cache.3. The system of claim 1 wherein said list of inodes comprises a first plurality of inodes for one or more directories and a second plurality of inodes for one or more files.4. The system of claim 3 wherein each directory of said one or more directories is a specially formatted file.5. The system of claim 3 wherein said first plurality of inodes for one or more directories are selected for write allocation before said second plurality of inodes for one or more files.6. The system of claim 5 wherein said second plurality of inodes are write allocated on a least recently used basis.7. A memory storing information including instructions, the instructions executable by a processor to allocate files in a file system, the instructions comprising the steps of: (a) selecting an inode for a RAID array having at least one dirty block from a list of inodes having dirty blocks; (b) write allocating blocks of data referenced by said inode to a storage device in said RAID array wherein said storage device comprises a plurality of storage blocks and adding said blocks of data to a list of writeable blocks of data for said storage device; (c) determining if all nodes have been processed, and when all of said inodes in said list of inodes have not been processed, continue repeating steps a-b; and, (d) flushing all unwritten stripes to said RAID array; wherein said blocks of data referenced by said inode are write allocated in a depth-first, post visit traversal of said blocks of data. 8. The memory of claim 7 wherein said blocks of data are stored in a cache.9. The memory of claim 7 wherein said list of inodes comprises a first plurality of inodes for one or more directories and a second plurality of inodes for one or more files.10. The memory of claim 9 wherein each directory of said one or more directories is a specially formatted file.11. The memory of claim 9 wherein said first plurality of inodes for one or more directories are selected for write allocation before said second plurality of inodes for one or more files.12. The memory of claim 11 wherein said second plurality of inodes are write allocated on a least recently used basis.13. A system for allocating files in a file system comprising: (a) means for selecting an inode for a RAID array having at least one dirty block from a list of inodes having dirty blocks; (b) means for write allocating blocks of data referenced by said inode to a storage device in said RAID array wherein said storage device comprises a plurality of storage blocks and adding said blocks of data to a list of writeable blocks of data for said storage device; (c) means for determining if all nodes have been processed, and when all of said inodes in said list of inodes have not been processed, continue repeating operations performed by means a and b; and, (d) means for flushing all unwritten stripes to said RAID array; wherein said blocks o f data referenced by said inode are write allocated in a depth-first, post visit traversal of said blocks of data. 14. A method of allocating files in a file system, comprising the steps of: (a) selecting an inode for a RAID array having at least one dirty block from a list of inodes having dirty blocks; (b) write allocating blocks of data referenced by said inode to a storage device in said RAID array wherein said storage device comprises a plurality of storage blocks and adding said blocks of data to a list of writeable blocks of data for said storage device; (c) determining if all nodes have been processed, and when all of said inodes in said list of inodes have not been processed, continue repeating steps a-b; and (d) flushing all unwritten stripes to said RAID array; wherein said blocks of data referenced by said inode are write allocated in a depth-first, post visit traversal of said blocks of data. 15. The method of claim 14 wherein said blocks of data are stored in a cache.16. The method of claim 14 wherein said list of inodes comprises a first plurality of inodes for one or more directories and a second plurality of inodes for one or more files.17. The method of claim 16 wherein each directory of said one or more directories is a specially formatted file.18. The method of claim 16 wherein said first plurality of inodes for one or more directories are selected for write allocation before said second plurality of inodes for one or more files.19. The method of claim 18 wherein said second plurality of inodes are write allocated on a least recently used basis.