Patent Application
20070106869
Typical computer systems include a file system for storing and accessing files. In addition to storing system files (operating system files, device driver files, etc.), the file system provides storage and access of user data files. If any of these files (system files and/or user files) contain critical data, then it becomes advantageous to employ a data backup scheme to ensure that critical data are not lost if a file storage device fails. One data backup scheme that is commonly employed is mirroring. Mirroring involves maintaining two or more copies of a file, where each copy of the file is located on a separate file storage device (e.g., a local hard disk, a networked hard disk, a network file server, etc.).
When one or more file storage devices fails for any length of time, the file storage device(s) may become unsynchronized. However, when employing a mirroring scheme, it is of critical importance to ensure that the mirrors are synchronized (i.e., that the contents of each mirror are the same). If a mirror becomes unsynchronized, the simplest recovery scheme involves copying all of the data from a synchronized mirror to the unsynchronized mirror. However, copying all data from one file storage device to another file storage device may take a long time and reduce performance of the file storage devices significantly during the resynchronization process.
Alternatively, dirty region logging (DRL) may be used to facilitate resynchronization. DRL involves dividing each mirror into a number of “regions.” Depending on the implementation, the region may be as small as a single disk sector or larger than 256 kilobytes (KB). Prior to modifying the content of a region—for example, when there is a write operation on data within the region—a DRL entry for the region is created in the DRL. In most cases, the DRL entry merely identifies the region where the modification will be attempted. If the region is modified successfully, then the DRL entry is cleared. If the region is not modified successfully, then the DRL entry remains in the DRL. Thus, during a resynchronization process, the DRL may be used to identify which specific regions require resynchronization, rather than resynchronizing the entire file storage device.
Dirty region logging may be more time-efficient then resynchronizing an entire file storage device. However, it also includes system overhead with each modification to a region, since the DRL must be updated prior to each modification to the region. Clearly, this overhead increases with smaller region sizes. Conversely, if the regions are large, there may be significant overhead involved in resynchronizing an entire region, even though only a single disk sector in that region may have been modified.
In general, in one aspect, the invention relates to a method for storing a block in a file system. The method comprises attempting to store the block at a location on a disk, wherein the block is associated with a birth time and is one selected from a group consisting of a data block and a first indirect block, and updating a dirty time log (DTL) with the birth time if the block is not stored successfully on the disk.
In general, in one aspect, the invention relates to a system. The system comprises a disk configured to store a block, wherein the block is associated with a birth time and is one selected from a data block and a first indirect block, a storage pool allocator configured to store the block on the disk, and a dirty time log (DTL) associated with a file system and configured to store the birth time if the block is not stored successfully.
Other aspects of the invention will be apparent from the following description and the appended claims.
Specific embodiments of the invention will now be described in detail with reference to the accompanying figures. Like elements in the various figures are denoted by like reference numerals for consistency.
In the following detailed description of embodiments of the invention, numerous specific details are set forth in order to provide a more thorough understanding of the invention. However, it will be apparent to one of ordinary skill in the art that the invention may be practiced without these specific details. In other instances, well-known features have not been described in detail to avoid unnecessarily complicating the description.
In general, embodiments of the invention provide a method and system to store a block. Specifically, embodiments of the invention attempt to store a data block or indirect block on a disk and update a dirty time log (DTL) if the attempt is not successful.
The operating system (103) typically interfaces with the file system (100) via a system call interface (102). The operating system (103) provides operations (101) for users to access files within the file system (100). These operations (101) may include read, write, open, close, etc. In one embodiment of the invention, the file system (100) is an object-based file system (i.e., both data and metadata are stored as objects). More specifically, the file system (100) includes functionality to store both data and corresponding metadata in the storage pool (108). Thus, the aforementioned operations (101) provided by the operating system (103) correspond to operations on objects.
More specifically, in one embodiment of the invention, a request to perform a particular operation (101) (i.e., a transaction) is forwarded from the operating system (103), via the system call interface (102), to the DMU (104). In one embodiment of the invention, the DMU (104) translates the request to perform an operation on an object directly to a request to perform a read or write operation at a physical location within the storage pool (108). More specifically, the DMU (104) represents the objects as data blocks and indirect blocks as described in
In one embodiment of the invention, the storage pool (108) includes one or more physical disks (disks (110A, 110N)). Further, in one embodiment of the invention, the storage capacity of the storage pool (108) may increase and decrease dynamically as physical disks are added and removed from the storage pool (108). In one embodiment of the invention, the storage space available in the storage pool (108) is managed by the SPA (106).
As shown in
As an alternative to storing the actual time that the disk (e.g., 110A, 110N) was offline, each I/O request issued to the storage pool (108) may be associated with a transaction group number. Thus, instead of storing the actual time the disk (e.g., 110A, 110N) was offline, the DTL (e.g., 115A, 115N) may store the transaction group number(s) during which the disk (e.g., 110A, 110N) was offline. More specifically, the DTL (e.g., 115A, 115N) may store the transaction group number(s) during which I/O requests were not successfully completed for the disk (e.g., 110A, 110N). Accordingly, the use of “birth time” in the following descriptions may refer to either a time or a transaction group number.
Those skilled in the art will appreciate that a single DTL may be maintained for the entire storage pool (108), where each entry identifies the particular disk (e.g., 110A, 110N) and when data could not be written to the disk because, for example, the disk (e.g., 110A, 110N) was offline. In addition, those skilled in the art will appreciate that the DTLs (e.g., 115A, 115N) may be stored in any location in the system (e.g., in the storage pool (108), in memory, etc.). In one embodiment of the invention, the SPA (106) manages the DTL (e.g., 115A, 115N) (i.e., updates the contents of the DTL (e.g., 115A, 115N), etc.).
As noted above, the SPA (106) receives transactions from the DMU (104). More specifically, the I/O management module (200), within the SPA (106), receives transactions from the DMU (104) and groups the transactions into transaction groups, in accordance with one embodiment of the invention. The compression module (201) provides functionality to compress larger logical blocks (i.e., data blocks and indirect blocks) into smaller segments, where a segment is a region of physical disk space. For example, a logical block size of 8 KB may be compressed to a size of 2 KB for efficient storage. Further, in one embodiment of the invention, the encryption module (202) provides various data encryption algorithms. The data encryption algorithms may be used, for example, to prevent unauthorized access. In one embodiment of the invention, the checksum module (203) includes functionality to calculate a checksum for data (i.e., data stored in a data block) and metadata (i.e., data stored in an indirect block) within the storage pool (108). The checksum may be used, for example, to ensure data has not been corrupted.
As discussed above, the SPA (106) provides an interface to the storage pool (108) and manages allocation of storage space within the storage pool (108). More specifically, in one embodiment of the invention, the SPA (106) uses the metaslab allocator (204) to manage the allocation of storage space in the storage pool (108).
In one embodiment of the invention, the storage space in the storage pool (108) is divided into contiguous regions of data, i.e., metaslabs. The metaslabs may in turn be divided into segments (i.e., portions of the metaslab). The segments may all be the same size, or alternatively, may be a range of sizes. The metaslab allocator (204) includes functionality to allocate large or small segments to store data blocks and indirect blocks. In one embodiment of the invention, allocation of the segments within the metaslabs is based on the size of the blocks within the I/O requests. That is, small segments are allocated for small blocks, while large segments are allocated for large blocks. The allocation of segments based on the size of the blocks may allow for more efficient storage of data and metadata in the storage pool (108) by reducing the amount of unused space within a given metaslab. Further, using large segments for large blocks may allow for more efficient access to data (and metadata) by reducing the number of DMU (104) translations and/or reducing the number of I/O operations. In one embodiment of the invention, the metaslab allocator (204) may include a policy that specifies a method to allocate segments.
As noted above, the storage pool (108) is divided into metaslabs, which are further divided into segments. Each of the segments within the metaslab may then be used to store a data block (i.e., data) or an indirect block (i.e., metadata).
In contrast to the root block (300), indirect blocks (302, 304, 306) and data blocks (308, 310, 312, 314) may be located anywhere in the storage pool. In one embodiment of the invention, the root block (300) and each block pointer (302A, 302B, etc.) include data as shown in the expanded diagram of block pointer (302B). One skilled in the art will appreciate that data blocks (308, 310, 312, 314) do not include this information; rather, data blocks (308, 310, 312, 314) contain actual data of files within the file system (e.g., 100 in
In one embodiment of the invention, each block pointer (e.g., 302A, 302B, etc.) includes a metaslab ID (318), an offset (320) within the metaslab, a birth time (322) of the block (i.e., data block or indirect block) referenced by the block pointer, and a checksum (324) of the data stored in the referenced block. In one embodiment of the invention, the metaslab ID (318) and offset (320) are used to determine the location of the referenced block in the storage pool. The metaslab ID (318) identifies a particular metaslab. More specifically, the metaslab ID (318) may identify the particular disk within the storage pool upon which the metaslab resides and the location on the disk where the metaslab begins. The offset (320) may then be used to reference a particular segment within the metaslab. In one embodiment of the invention, the data within the segment referenced by the particular metaslab ID (318) and offset (320) may correspond to either a data block or an indirect block. If the data corresponds to an indirect block, then the metaslab ID (318) and offset (320) within a block pointer in the indirect block are extracted and used to locate a subsequent data block or indirect block. The tree may be traversed in this manner to eventually retrieve a requested data block.
In one embodiment of the invention, copy-on-write transactions are performed for every data write request to a file. Specifically, all write requests cause new segments to be allocated for the modified data. Therefore, the retrieved data blocks and indirect blocks are never overwritten (until a modified version of the data block and indirect block is committed). More specifically, the DMU writes out all the modified data blocks in the tree to unused segments within the storage pool. Subsequently, the DMU writes out the corresponding block pointers (within indirect blocks) to unused segments in the storage pool. In one embodiment of the invention, fields (i.e., metaslab ID, offset, birth, checksum) for the corresponding block pointers are populated by the DMU prior to sending an I/O request to the SPA. The indirect blocks containing the block pointers are typically written one level at a time. To complete the copy-on-write transaction, the SPA issues a single write that atomically changes the root block to reference the indirect blocks referencing the modified data block.
In one embodiment of the invention, the transaction includes one or more data blocks and/or one or more indirect blocks. As noted above, the file system is stored on disk using a hierarchical structure including data blocks and indirect blocks. Thus, for a given transaction, the initial transaction group includes I/O requests for the data blocks to be written to disk, while subsequent transaction groups include I/O requests corresponding to the indirect blocks containing one or more block pointers. Accordingly, the transaction group referenced in ST104 includes I/O requests to write data blocks.
Continuing with the discussion of
Continuing with the discussion of
Next, the SPA receives and subsequently attempts to write the indirect blocks into the storage pool (ST116). Those skilled in the art will appreciate that if an indirect block is not written successfully, then the birth time of the block may be stored in a dirty time log (DTL), indicating that the write attempt was not successful. Further, those skilled in the art will appreciate that the birth time(s) stored in the DTL may be equivalent to the birth time(s) that would have been associated with a successful write to disk. Further, those skilled in the art will appreciate that if the write attempt fails, an attempt may be made to store the indirect block at another location. If the indirect block is stored in another location, then the DTL may not be updated.
Continuing with the discussion of
Those skilled in the art will appreciate that updating a DTL if a write attempt fails, as described above, offers fast performance and/or low system overhead, because the DTL is only updated if a write fails. Further, the use of birth times in the DTL minimizes the amount of storage space required to maintain the DTL. In addition, if many writes fail with the same birth time, the DTL only has to be updated once.
If the birth time is not in the DTL, then the location (e.g., the metaslab ID and offset) is subsequently used to obtain the block (i.e., the data block or the indirect block) (ST140). Then a determination is made about whether there are more indirect blocks to read (ST142). That is, if the retrieved block is a data block, then there are no more indirect blocks to read, so the data are extracted from the data block and presented to the process requesting the data (ST146). Alternatively, if the retrieved block is not a data block, then the birth time and location of the next block (stored within a block pointer within the retrieved block) is obtained (ST144). Steps ST134 through ST144 are subsequently repeated until either data cannot be retrieved (i.e., ST136, as discussed below) or the data block is encountered (i.e., ST146).
If the birth time is in the DTL, then a determination is made about whether an alternate location of the block is available (ST136). In one embodiment of the invention, the alternate location of the block may correspond to a location of a copy of the block stored on another disk (e.g., a mirror). If an alternate location of the block is not available, then the process ends and the data cannot be retrieved. If an alternate location is available, then the alternate location of the block as well as the birth time of the block at the alternate location are obtained (ST138). The process then proceeds to ST134.
Those skilled in the art will appreciate that the method of
The invention may be implemented on virtually any type of computer regardless of the platform being used. For example, as shown in
Further, those skilled in the art will appreciate that one or more elements of the aforementioned computer system (700) may be located at a remote location and connected to the other elements over a network. Further, the invention may be implemented on a distributed system having a plurality of nodes, where each portion of the invention (e.g., operating system, file system, system call interface, DMU, SPA, storage pool, disk, dirty time log, metaslab allocator, I/O management module, compression module, encryption module, checksum module, root block, data block, indirect block, etc.) may be located on a different node within the distributed system. In one embodiment of the invention, the node corresponds to a computer system. Alternatively, the node may correspond to a processor with associated physical memory. The node may alternatively correspond to a processor with shared memory and/or resources. Further, software instructions to perform embodiments of the invention may be stored on a computer readable medium such as a compact disc (CD), a diskette, a tape, a file, or any other computer readable storage device.
While the invention has been described with respect to a limited number of embodiments, those skilled in the art, having benefit of this disclosure, will appreciate that other embodiments can be devised which do not depart from the scope of the invention as disclosed herein. Accordingly, the scope of the invention should be limited only by the attached claims.
This application claims benefit of U.S. Provisional Application Ser. No. 60/734,023 filed on Nov. 4, 2005, entitled “DIRTY TIME LOGGING AND RESILVERING” in the names of William H. Moore and Jeffrey S. Bonwick. This application is related to copending U.S. patent application Ser. No. ______, filed on and entitled “Method and System for Dirty Time Log Directed Resilvering,” copending U.S. patent application Ser. No.______ , filed on______and entitled “Method and System for Metadata-Based Resilvering,” and copending U.S. patent application Ser. No.______, filed on______and entitled “Method and System for Pruned Resilvering Using a Dirty Time Log,” the entire contents of which are incorporated herein by reference. All the referenced applications are co-owned by the same assignee. The present application contains subject matter that may be related to the subject matter in the following U.S. patent applications, which are all assigned to a common assignee: “Method and Apparatus for Self-Validating Checksums in a File System” (application Ser. No. 10/828,573) filed on Apr. 24, 2004; “Method and Apparatus for Dynamic Striping” (application Ser. No. 10/828,677) filed on Apr. 21, 2004; “Method and Apparatus for Vectored Block-Level Checksum for File System Data Integrity” (application Ser. No. 10/828,715) filed on Apr. 21, 2004; “Method and Apparatus for Identifying Tampering of Data in a File System” (application Ser. No. 10/853,874) filed on May 26, 2004; “Method and System for Detecting and Correcting Data Errors Using Checksums and Replication” (application Ser. No. 10/853,837) filed on May 26, 2004; “Method and System for Detecting and Correcting Data Errors Using Data Permutations” (application Ser. No. 10/853,870) filed on May 26, 2004; “Method and Apparatus for Compressing Data in a File System” (application Ser. No. 10/853,868) filed on May 26, 2004; “Gang Blocks” (application Ser. No. 10/919,878) filed on Aug. 17, 2004; “Method and Apparatus for Enabling Adaptive Endianness” (application Ser. No. 10/919,886) filed on Aug. 17, 2004; and “Automatic Conversion of All-Zero Data Storage Blocks into File Holes” (application Ser. No. 10/853,915) filed on May 26, 2004.
| Number | Date | Country | |
|---|---|---|---|
| 60734023 | Nov 2005 | US |
