×

Detecting deadlock in a cluster environment using big data analytics

  • US 10,528,400 B2
  • Filed: 06/05/2017
  • Issued: 01/07/2020
  • Est. Priority Date: 06/05/2017
  • Status: Active Grant
First Claim
Patent Images

1. A method for detecting deadlock in a distributed computing environment, by a processor device, comprising:

  • detecting potential deadlocks between resources of nodes in a computing cluster by determining resource reverse pairs of the resources for each transaction from trace or log files generated from additional, separate transactions using data analytics;

    wherein the potential deadlocks are identified offline, prior to any execution of each transaction, by matching a global or local resource between the nodes in sub-transactions of each transaction as recursively identified from a transaction resource chain having information of an order of the resources to be used during each transaction; and

    wherein the reverse resource pairs comprise those of the resources determined to collide when being taken by one of the sub-transactions while simultaneously being attempted to be taken by another of the sub-transactions as identified in the transaction resource chain;

    mitigating an occurrence of the potential deadlock by delaying the one of the sub-transactions until the respective resources used during the one of the sub-transactions are freed or restructuring the order of the resources to be used during the one of the sub-transactions;

    defining namespaces for each of the transactions, the sub-transactions, and the resources used by the transactions and sub-transactions;

    wherein the namespaces are recorded in the log or trace files; and

    extracting information associated with a respective transaction and a resource order of the resources used for the respective transaction from the log or trace files.

View all claims
  • 1 Assignment
Timeline View
Assignment View
    ×
    ×