Storage and access of continuous media files indexed as lists of raid stripe sets associated with file names

• US 5,974,503 A
• Filed: 05/05/1997
• Issued: 10/26/1999
• Est. Priority Date: 04/25/1997
• Status: Expired due to Term

• Current Assignee: EMC Corporation (Dell Technologies Inc.)
• Original Assignee: EMC Corporation (Dell Technologies Inc.)
• Inventors: Venkatesh, Dinesh, Forecast, John, Duso, Wayne W.
• Primary Examiner(s): Cabeca, John W.
• Assistant Examiner(s): GOODARZI, NASSER MOAZZAMI
• Application Number:
  US08/851,509
• Time in Patent Office: 904 Days
• Field of Search: 711/114, 371/51.1, 371/40.4, 395/182.04, 395/182.05
• US Class Current: 711/114
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/1471 involving logging of persistent data for recovery
• CPC Class Codes: G06F 11/2023 Failover techniques
• CPC Class Codes: G06F 11/2035 without idle spare hardware
• CPC Class Codes: G06F 11/2046 where the redundant components share persistent storage G06F11/2043 takes precedence
• CPC Class Codes: G06F 2211/1009 Cache, i.e. caches used in RAID system with parity
• CPC Class Codes: G06F 2211/1014 Compression, i.e. RAID systems with parity using compression techniques
• CPC Class Codes: H04N 21/2318 using striping
• CPC Class Codes: H04N 21/232 Content retrieval operation locally within server, e.g. reading video streams from disk arrays storage management G06F3/0604; details of querying and searching of video data from a database G06F16/739
• CPC Class Codes: H04N 21/2387 Stream processing in response to a playback request from an end-user, e.g. for trick-play
• CPC Class Codes: H04N 21/2396 characterized by admission policies admission control, resource allocation in open networks H04L12/5692; arrangements for network security using user profiles for access control H04L63/102; access security in wireless networks H04W12/08
• CPC Class Codes: H04N 21/2404 Monitoring of server processing errors or hardware failure error or fault detection G06F11/07; monitoring in general G06F11/30

Abstract

A continuous media file is comprised of stripe sets over disk drives in one or more RAID sets. In a preferred embodiment, the RAID set includes n disk drives. The data storage of each disk drive in the RAID set is partitioned into an integer number m of hyper-volumes, and the parity is stored in one hyper-volume of each of m disk drives in the RAID set. The stripe set includes a series of transfer units of data in respective ones of the disk drives. Each transfer unit includes an integer number j of data blocks, and each hyper-volume includes an integer number k of transfer units. Each stripe set includes (m)(n-1) transfer units of data. The transfer units of the RAID set are allocated for the storage of continuous media data in a right-to-left and then top-to-bottom order in which the transfer units appear in an m row by n column matrix in which the rows of the matrix represent parity groups of hyper-volumes in the disk drives and the columns of the matrix represent storage in the respective disk drives. At most one write access to each parity hyper-volume need be performed during write access to a stripe set. Parity changes for the data being written are accumulated in non-volatile memory, and written to the RAID set after completion of the writing of the data.

Claims

1. A method of striping a sequence of continuous media data across a plurality of n disk drives in a RAID set storing data and associated parity across the disk drives, wherein the data storage of each disk drive in the RAID set is partitioned into an integer number m of hyper-volumes, parity is distributed among the disk drives in the RAID set and is stored in at least one hyper-volume of each of m disk drives in the RAID set, and transfer units of data storage in the n disk drives are associated with the sequence of continuous media data in a right-to-left and then top-to-bottom order in which the transfer units appear in an m row by n column matrix in which the rows of the matrix represent parity groups of hyper-volumes in the disk drives and the columns of the matrix represent the data storage in the respective n disk drives in the RAID set, which further includes allocating data storage in the RAID set for the stream of continuous media data in fixed size stripe sets, each stripe set comprising a segment of the sequence of continuous media data distributed across each of the disk drives and each of the parity groups of hyper-volumes in the disk drives,

wherein the transfer unit has a predetermined size, and each stripe set includes (m)(n-1) transfer units of data.

3. The method as claimed in claim 1, which includes computing parity for each of the parity groups by computing parity changes for different ones of the disk drives due to storing of the sequence of continuous media data in the disk drives, reading the parity once from each of the parity groups, applying the parity changes for the different ones of the disk drives to the parity read once for each of the parity groups to produce modified parity, and writing the modified parity back to said each of the parity groups.

2. A method of striping a sequence of continuous media data across a plurality of n disk drives in a RAID set storing data and associated parity across the disk drives, wherein the data storage of each disk drive in the RAID set is partitioned into an integer number m of hyper-volumes, parity is distributed among the disk drives in the RAID set and is stored in at least one hyper-volume of each of m disk drives in the RAID set, and transfer units of data storage in the n disk drives are associated with the sequence of continuous media data in a right-to-left and then top-to-bottom order in which the transfer units appear in an m row by n column matrix in which the rows of the matrix represent parity groups of hyper-volumes in the disk drives and the columns of the matrix represent the data storage in the respective n disk drives in the RAID set, which further includes allocating data storage in the RAID set for the stream of continuous media data in fixed size stripe sets, each stripe set comprising a segment of the sequence of continuous media data distributed across each of the disk drives and each of the parity groups of hyper-volumes in the disk drives, and which includes associating a name of a continuous media data file with a list of the stripe sets.

4. A method of accessing a continuous media data file including continuous media data stored in stripe sets in a RAID set of disk drives, said method comprising the steps of:

a) accessing a directory of file names of continuous media data files to locate a list of the stripe sets associated with a name of the continuous media data file;

b) accessing the list of stripe sets to determine a sequence of the stripe sets storing the continuous media data; and

c) accessing the RAID set of disk drives to retrieve the sequence of stripe sets.

5. The method as claimed in claim 4, wherein the list of stripe sets includes at least one entry specifying a starting stripe set number and an ending stripe set number, and said method includes accessing the RAID set of disk drives to retrieve a stripe set identified by the starting stripe set number, to retrieve stripe sets identified by stripe set numbers between the starting stripe set number and the ending stripe set number, and to retrieve a stripe set identified by the ending stripe set number.

6. The method as claimed in claim 4, wherein the stripe set is comprised of a series of transfer units in disk drives of the RAID set, and said method includes the step of sequentially advancing an index and accessing a look-up table with the index in order to determine a disk drive in the RAID set containing a next transfer unit to be accessed.

7. A data storage system comprising:

a plurality of n disk drives arranged in a RAID set; and

a controller coupled to the disk drives for accessing a sequence of continuous media data stored in the disk drives;

wherein the sequence of continuous media data and associated parity is striped across the disk drives in the RAID set, the data storage of each disk drive in the RAID set is partitioned into an integer number m of hyper-volumes, parity is distributed among the disk drives in the RAID set and is contained in at least one hyper-volume of each of the m disk drives in the RAID set, and transfer units of data storage in the n disk drives are associated with the sequence of continuous media data in a right-to-left and then top-to-bottom order in which the transfer units appear in an m row by n column matrix in which the rows of the matrix represent parity groups of hyper-volumes in the disk drives and the columns of the matrix represent the data storage in the respective n disk drives in the RAID set,

wherein data storage in the RAID set for the stream of continuous media data is allocated in fixed size stripe sets, each stripe set comprising a segment of the sequence of continuous media data distributed across each of the disk drives and each of the parity groups of hyper-volumes in the disk drives, and wherein the transfer unit has a predetermined size, and each stripe set includes (m)(n-1) transfer units of data.

9. The data storage system as claimed in claim 7, wherein the controller is programmed to compute parity for each of the parity groups by computing parity changes for different ones of the disk drives due to storing of the sequence of continuous media data in the disk drives, to read the parity once from each of the parity groups, to apply the parity changes for the different ones of the disk drives to the parity read once for each of the parity groups to produce modified parity, and to write the modified parity back to said each of the parity groups.

8. A data storage system comprising:

a plurality of n disk drives arranged in a RAID set; and

a controller coupled to the disk drives for accessing a sequence of continuous media data stored in the disk drives;

wherein the sequence of continuous media data and associated parity is striped across the disk drives in the RAID set, the data storage of each disk drive in the RAID set is partitioned into an integer number m of hyper-volumes, parity is distributed among the disk drives in the RAID set and is contained in at least one hyper-volume of each of the m disk drives in the RAID set, and transfer units of data storage in the n disk drives are associated with the sequence of continuous media data in a right-to-left and then top-to-bottom order in which the transfer units appear in an m row by n column matrix in which the rows of the matrix represent parity groups of hyper-volumes in the disk drives and the columns of the matrix represent the data storage in the respective n disk drives in the RAID set, wherein data storage in the RAID set for the stream of continuous media data is allocated in fixed size stripe sets, each stripe set comprising a segment of the sequence of continuous media data distributed across each of the disk drives and each of the parity groups of hyper-volumes in the disk drives, and wherein the controller has a directory associating a name of a continuous media data file with a list of the stripe sets.

10. A data storage system comprising:

a plurality of disk drives arranged in a RAID set; and

a controller coupled to the disk drives for accessing a sequence of continuous media data stored in stripe sets in a RAID set of disk drives, the controller having a directory of continuous media data files and lists of the stripe sets associated with each of the continuous media data files, the controller being programmed to respond to a request to access a specified continuous media data file by accessing the directory to find a list of stripe sets associated with the specified continuous media data file, to access the list of stripe sets to determine a sequence of the stripe sets storing the continuous media data, and to access the RAID set of disk drives to retrieve the sequence of stripe sets.

11. A data storage system comprising:

a plurality of disk drives arranged in a RAID set; and

a controller coupled to the disk drives for accessing a sequence of continuous media data stored in stripe sets in a RAID set of disk drives, the controller having a directory of continuous media data files and lists of the stripe sets associated with each of the continuous media data files, the controller being programmed to respond to a request to access a specified continuous media data file by accessing the directory to find a list of stripe sets associated with the specified continuous media data file, to access the list of stripe sets to determine a sequence of the stripe sets storing the continuous media data, and to access the RAID set of disk drives to retrieve the sequence of stripe sets, wherein the list of stripe sets includes at least one entry specifying a starting stripe set number and an ending stripe set number, and the controller is programmed to access the RAID set of disk drives to retrieve a stripe set identified by the starting stripe set number, to retrieve stripe sets identified by stripe set numbers between the starting stripe set number and the ending stripe set number, and to retrieve a stripe set identified by the ending stripe set number.

12. A data storage system comprising:

a plurality of disk drives arranged in a RAID set; and

a controller coupled to the disk drives for accessing a sequence of continuous media data stored in stripe sets in a RAID set of disk drives, the controller having a directory of continuous media data files and lists of the stripe sets associated with each of the continuous media data files, the controller being programmed to respond to a request to access a specified continuous media data file by accessing the directory to find a list of stripe sets associated with the specified continuous media data file, to access the list of stripe sets to determine a sequence of the stripe sets storing the continuous media data, and to access the RAID set of disk drives to retrieve the sequence of stripe sets, and wherein the stripe set is comprised of a series of transfer units in disk drives of the RAID set, and the controller is programmed to sequentially advance an index and access a look-up table with the index in order to determine a disk drive in the RAID set containing a next transfer unit to be accessed.