Failover system and method for cluster environment

• US 7,139,930 B2
• Filed: 07/25/2005
• Issued: 11/21/2006
• Est. Priority Date: 08/09/2001
• Status: Expired due to Term

• Current Assignee: Dell Products LP (Dell Technologies Inc.)
• Original Assignee: Dell Products LP (Dell Technologies Inc.)
• Inventors: Rooholamini, Mohamad Reza, Hsieh, Jenwei, Mashayekhi, Victor
• Primary Examiner(s): Baderman, Scott
• Assistant Examiner(s): Bonura, Timothy M.
• Application Number:
  US11/188,429
• Publication Number:
  US 20050268156A1
• Time in Patent Office: 484 Days
• Field of Search: 714 3- 4
• US Class Current: 714/4.11
• CPC Class Codes: G06F 11/2028 eliminating a faulty processor or activating a spare
• 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: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

A failover method and system is provided for a computer system having at least three nodes operating as a cluster. One method includes the steps of detecting failure of one node, determining the weight of at least two surviving nodes, and assigning a failover node based on the determined weights of the surviving nodes. Another method includes the steps detecting failure of one node and determining the time of failure, and assigning a failover node based in part on the determined time of failure. This method may also include the steps of determining a time period during which nodes in the cluster are heavily utilized, and assigning a failover node that is not heavily utilized during that time period.

Claims

1. A failover method for a computer system having at least three nodes operating as a cluster, said method comprising the steps of:

detecting a failure of one of the nodes;

determining a time of failure of the failed node;

for a second node in the cluster, determining a time period during which the second node is heavily utilized;

preventing the second node from being assigned as a failover node during the time period during which it is heavily utilized;

assigning a failover node, other than the second node, taking into consideration the determined time of failure of the failed node; and

assigning applications running on the failed node to the failover node.

2. The failover method according to claim 1, further comprising the steps of:

for at least the failed node, determining in advance of failure a time during which at least one application running on the failed node is heavily utilized; and

following the failure of the failed node, if failure occurs during the determined time period, then assigning a failover node for the at least one application before assigning a failover node for any of the other applications of the failed node.

3. A cluster computer system, comprising:

at least three nodes, wherein the nodes are computer systems operating as a cluster, wherein the cluster computer system is capable of determining a time of failure of one of the nodes, and implementing a failover policy in which, following a failure of the failed node, a failover node is selected based in part on the determined time, and wherein the cluster computer system is capable of identifying a time period which at least one of the nodes is heavily utilized, preventing the heavy utilized node from being assigned as a failover node during the identified time period, assigning a failover node, other than the heavily utilized node, taking into consideration the determined time of failure of the failed node and assigning applications running on the failed node to the failover node.

4. The cluster computer system of claim 3, wherein the cluster computer system is further capable of determining a time period during which at least one application on the failed node is heavily utilized, and upon failure of the failed node, if the failure occurs during the determined time, then failing over the at least one application before failing over any of the other applications of the failed node.

5. A failover method for a computer system having at least three nodes operating as a cluster, the method comprising the steps of:

determining a time period during which each of the nodes is heavily utilized;

detecting a failure of one of the nodes;

determining a time of failure of the failed node;

preventing failover of applications running on the failed node to a second node if the failure of the failed node occurs during a time period in which the second node is heavily utilized;

assigning a failover node, other than the second node, taking into consideration the determined time of failure of the failed node; and

assigning applications running on the failed node to the failover node.