Asynchronous write queues, reconstruction and check-pointing in disk-mirroring applications

• US 6,260,125 B1
• Filed: 12/09/1998
• Issued: 07/10/2001
• Est. Priority Date: 12/09/1998
• Status: Expired due to Term

• Current Assignee: Teradata US, Inc. (Teradata Corporation)
• Original Assignee: NCR Corporation
• Inventors: McDowell, Steven R.
• Primary Examiner(s): Verbrugge, Kevin
• Application Number:
  US09/207,935
• Time in Patent Office: 944 Days
• Field of Search: 711/111, 711/112, 711/114, 711/161, 711/162, 711/152, 714/6, 707/10, 707/204
• US Class Current: 711/162
• CPC Class Codes: G06F 11/1471 involving logging of persistent data for recovery
• CPC Class Codes: G06F 11/2074 Asynchronous techniques
• CPC Class Codes: G06F 11/2082 Data synchronisation

Abstract

An asynchronous disk mirroring system for use within a network computer system. The disk mirroring system includes a first storage volume connected to receive write requests from the computer system; a write queue connected to also receive the write requests directed to the first storage volume; and a second storage volume connected to the write queue to receive the write requests. The write queue operates to delay the time of receipt of the write requests by the second storage volume. Write requests are passed through the write queue, comprising a plurality of serially connected write buffers, in a first-in, first-out (FIFO) order. A log file connected to receive the delayed write requests from the write queue is also included in the disk mirroring system to provide for log-based mirror reconstruction and check-pointing of the mirrored volumes.

Claims

1. A computer system comprising:

a first storage volume connected to receive write requests from said computer system;

a write queue connected to receive said write requests directed to said first storage volume;

a second storage volume connected to said write queue to receive said write requests, said write queue operating to delay the time of receipt of said write requests by said second storage volume;

a mechanism for determining whether said write queue is full beyond a “

high water”

mark and blocking the processing of further write requests to said first storage volume and said write queue, and

a mechanism for determining whether said write queue is cleared below a “

low water”

mark following said blocking of the processing of further write requests to said first storage volume and said write queue, and resuming the processing of further write requests to said first storage volume and said write queue.

2. The computer system in accordance with claim 1, wherein:

said write queue comprises a plurality of serially connected write buffers, wherein said write requests pass through said write queue in a first-in, first-out (FIFO) sequence.

3. The computer system in accordance with claim 1, further comprising:

a log file connected to said write queue to receive said write requests.

4. A computer system comprising:

a first server computer including a first storage volume connected to receive write requests from said computer system;

a second server computer including a second storage volume;

a communications network connecting said first and second server computers;

a write queue connected to receive said write requests directed to said first storage volume, and connected to provide said write requests to said second storage volume through said communications network, said write queue operating to delay the time of receipt of said write requests by said second storage volume;

a mechanism for determining whether said write queue is full beyond a “

high water”

mark and blocking the processing of further write requests to said first storage volume and said write queue; and

a mechanism for determining whether said write queue is cleared below a “

low water”

mark following said blocking of the processing of further write requests to said first storage volume and said write queue, and resuming the processing of further write requests to said first storage volume and said write queue.

5. In a networked computer system including a first network node including a first storage volume, a second network node including a second storage volume, and a communications link connecting said first and second network nodes, apparatus for asynchronously mirroring write requests directed to said first storage volume to said second storage volume, said apparatus comprising:

a write queue connected to receive said write requests directed to said first storage volume, and connected to provide said write requests to said second storage volume through said communications link, said write queue operating to delay the time of receipt of said write requests by said second storage volume;

a mechanism for determining whether said write queue is full beyond a “

high water”

mark and blocking the processing of further write requests to said first storage volume and said write queue; and

a mechanism for determining whether said write queue is cleared below a “

low water”

mark following said blocking of the processing of further write requests to said first storage volume and said write queue, and resuming the processing of further write requests to said first storage volume and said write queue.

6. In a computer system including first storage volume, a method for mirroring write transactions directed to said first storage volume, said method comprising the steps of:

providing a second storage volume;

copying said write transactions directed to said first storage volume and providing said copied write transactions through a first-in first-out (FIFO) write queue to said second storage volume, said write queue operating to delay the time of receipt of said copied write requests by said second storage volume;

determining whether said write queue is full beyond a “

high water”

mark and blocking the processing of further write requests to said first storage volume and said write queue; and

determining whether said write queue is cleared below a “

low water”

mark following said blocking of the processing of further write requests to said first storage volume and said write queue, and resuming the processing of further write requests to said first storage volume and said write queue.

7. The method in accordance with claim 6, further comprising the step of:

copying storage device state information associated with said first storage volume and providing said copied storage device state information through said first-in first-out (FIFO) write queue to said second storage volume, said write queue operating to delay the time of receipt of said copied storage device state information by said second storage volume.