Loosely coupled mass storage computer cluster

• US 6,557,114 B2
• Filed: 09/18/2001
• Issued: 04/29/2003
• Est. Priority Date: 10/24/1995
• Status: Expired due to Term

• Current Assignee: SeaChange International Incorporated
• Original Assignee: SeaChange Technology, Inc.
• Inventors: Mann, Bruce E., Ywoskus, John A., McGrath, Edward J., Carlozzi, Michael D., Trasatti, Philip J.
• Primary Examiner(s): Iqbal, Nadeem
• Application Number:
  US09/954,830
• Publication Number:
  US 20020013916A1
• Time in Patent Office: 588 Days
• Field of Search: 714/6, 714/2, 714/4, 714/11, 714/13, 714/20, 714/25, 714/38, 714/43, 714/48, 711/114, 711/163, 709/201, 709/203
• US Class Current: 714/6.24
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/1088 Reconstruction on already foreseen single or plurality of spare disks

Abstract

A method and apparatus redundantly store data, in particular video data objects, in a distributed computer system having at least three processor systems, each processor system being connected in point to point two way channel interconnection with each other processor system. The data is stored in a redundant fashion both at the computer system level as well as the processor system level. Accordingly, the failure of a single processor does not adversely affect the integrity of the data. The computer system can also overlay a switching system connected in a ring fashion for providing a fault tolerance to the failure of a single connected processor system at the switch level. Accordingly, there results a fault tolerant data distribution system.

Claims

1. A method of retrieving a video object from a clustered video system that has a plurality of clustered members, the method comprising:

reading fragment files from the clustered members of the clustered system, with each clustered member maintaining a file system, and with the fragment files having a header block that allows the fragment files to be self-describing so that a cluster member can maintain its file system private and export access to the video object.

2. The method of claim 1 wherein each cluster member reads, writes, or deletes files from its local file system without disrupting other members of the cluster.

3. The method of claim 1 wherein the fragment files are distributed across at least some of the cluster members.

4. The method of claim 3 wherein each cluster member comprises:

a processor system including a CPU; and

local storage to store portions of the plurality of fragment files.

5. The method of claim 1 wherein the fragment files further comprise:

data blocks that store data that can render the video object on an output device.

6. The method of claim 1 wherein the fragment files further comprise

parity blocks that are calculated from data blocks.

7. The method of claim 1 wherein the fragment header comprises

a fragment count field that indicates the fragment count, a stripe size field that indicates a data object. stripe size, and a data object name field that indicates a data object name.

8. The method of claim 1 wherein each cluster member includes local storage for fragment files, and each cluster member reads, writes or deletes files from the cluster member local file system without disrupting other members of the cluster.

9. The method of claim 1 wherein the cluster system members execute a video application to manage access to the video object.

10. The method of claim 1 further comprising:

storing the video object on the clustered system; and

wherein storing further comprises;

defragmenting the video object into the plurality of fragment files; and

storing the file fragments across the video system.

11. A clustered video system for storing a video object, the clustered video system comprising:

a plurality of cluster members, each of the cluster members comprising;

local storage to store portions of the video objects as fragment files, the fragment files containing a header block and data blocks, the header block allowing the fragment files to be self-describing so that the cluster member can keep a private file system and export access only to the video object; and

a computer processor unit that executes a video application program to manage the access to the video object.

12. The clustered video system of claim 11 wherein each cluster member reads, writes or delete files from its local file system without disrupting the local file system of the other video cluster members.

13. The clustered video system of claim 11 wherein the fragment files are distributed across at least some of the cluster members.

14. The clustered video system of claim 11 wherein the cluster members are interconnected through a network.

15. The clustered video system of claim 11 wherein the fragment files further comprise parity blocks that are calculated from the data blocks.

16. The clustered video system of claim 11 wherein Huffman coding is used as a redundancy method to detect failure of a processor.

17. The clustered video system of claim 11 wherein the fragment header comprises:

a field that indicates the fragment count, a field that indicates video object stripe size, and a field that indicates the video object name.

18. The cluster video system of claim 11 wherein file system for the local storage of each cluster member is maintained for the named fragment files by the central processing unit of the cluster member.

19. The clustered video system of claim 11 wherein video object is stored as the plurality of fragment files across at least some of the members of the video cluster system.