Technique to monitor application behavior and tune replication performance

• US 7,334,062 B1
• Filed: 07/22/2003
• Issued: 02/19/2008
• Est. Priority Date: 07/22/2003
• Status: Expired due to Term

• Current Assignee: Veritas Technologies, LLC (Whitehouse Group Ltd.)
• Original Assignee: Symantec Operating Corporation (Gen Digital Inc.)
• Inventors: Gupta, Vikas Kailash, Agarwal, Ashish
• Primary Examiner(s): Kindred; Alford
• Assistant Examiner(s): Sorrell; Eron J.
• Application Number:
  US10/624,396
• Time in Patent Office: 1,673 Days
• Field of Search: None
• US Class Current: 710/52
• CPC Class Codes: G06F 9/5016 the resource being the memory

Abstract

A method and mechanism for modifying computing resources in response to application behavior. A computing system includes a replication component configured to replicate data storage from a first data volume to a second data volume. In addition, the replication component is configured to monitor application I/O characteristics and store related statistics. I/O characteristics may include size, concurrency, locality, and frequency. I/O characteristics which are stored, and guidelines for modifying system resources based on those characteristics, may be displayed for use by an administrator in tuning system resources. Periodically, or in response to detecting an event, the replication component may automatically access the statistics and modify the system resources used by the replication system to better accommodate the application'"'"'s behavior.

Claims

1. A computing system comprising:

an application configured to initiate write transactions;

a first storage device configured to store data corresponding to said write transactions;

a memory pool; and

a replicator component configured to;

monitor said write transactions;

allocate buffers from said memory pool for said write transactions; and

automatically modify system resources in response to I/O characteristics of said monitored write transactions, wherein modifying said system resources includes modifying a size of said memory pool;

wherein said application, first storage device, and replicator are within a first node of said system, and wherein said system includes a second node with a second storage device coupled to said first node, wherein said replicator component is further configured to convey said write transactions to said second node.

2. The computing system as recited in claim 1, wherein said replicator is further configured to record data indicative of said characteristics.

3. The computing system as recited in claim 2, wherein said replicator is further configured to:

provide said recorded characteristics for display;

provide guidelines to a user for modifying resources of said system; and

modify said resources based upon user input.

4. The computing system as recited in claim 1, further comprising a log volume, and wherein said replicator is further configured to store said write transactions in said log volume.

5. The computing system as recited in claim 1, wherein said second node includes a pool of buffers, each of which is configured to store a write transaction received from said first node, and wherein said replicator component is further configured to modify a size of said pool of buffers in said second node in response to said characteristics.

6. The computing system as recited in claim 5, wherein said replicator component is configured to access said recorded data responsive to detecting an event.

7. The system of claim 1, wherein said second node is configured to serve as a failover node if the first node fails.

8. A method comprising:

initiating write transactions in a first node of a computing system;

allocating buffers from a memory pool for said write transactions;

storing data corresponding to said write transactions in a first storage device of the first node;

monitoring said write transactions;

automatically modifying system resources in response to I/O characteristics of said monitored write transactions, wherein said modifying includes modifying a size of said memory pool; and

conveying said write transactions to a second node of the computing system and storing data corresponding to said write transactions in the second node.

9. The method as recited in claim 8, further comprising recording data indicative of said characteristics.

10. The method as recited in claim 9, further comprising:

providing said recorded statistics for display;

providing guidelines to a user for modifying resources of said system; and

modifying said resources based upon user input.

11. The method as recited in claim 8, further comprising storing said write transactions in a log volume.

12. The method as recited in claim 8, wherein said second node includes a pool of buffers, each of which is configured to store a write transaction received from said first node, and wherein said method further comprises modifying a size of said pool of buffers in said second node in response to said characteristics.

13. The method as recited in claim 12, further comprising accessing said recorded data responsive to detecting an event.

14. The method of claim 8, wherein said second node is configured to serve as a failover node if the first node fails.

15. A machine readable storage medium comprising program instructions, wherein said program instructions are executable to:

initiate write transactions in a first node of a computing system;

allocate buffers from a memory pool for said write transactions;

store data corresponding to said write transactions in a first storage device of the first node;

monitor said write transactions;

automatically modify system resources in response to I/O characteristics of said monitored write transactions, wherein modifying said system resources includes modifying a size of said memory pool; and

convey said write transactions to a second node of the computing system and store data corresponding to said write transactions in the second node.

16. The storage medium as recited in claim 15, wherein said program instructions are further executable to record data indicative of said characteristics.

17. The storage medium as recited in claim 15, wherein said program instructions are further executable to:

convey said write transactions from a first node to a buffer allocated from a pool of buffers within a second node; and

modify a size of said pool of buffers in said second node in response to said characteristics.

18. The storage medium of claim 15, wherein said second node is configured to serve as a failover node if the first node fails.