System and method for maintaining write order fidelity in a distributed environment
View Patent ↗Various systems and methods for maintaining write order fidelity in a distributed environment are disclosed. One method, which can be performed by each node in a cluster, involves associating a current sequence number with each of several write operations included in a set of independent write operations. In response to detecting that one of the write operations in the set is ready to complete, a new sequence number is selected, and that new sequence number is thereafter used as the current sequence number. None of write operations in the set is allowed to return to the application that initiated the write operations until the new sequence number has been advertised to each other node in the cluster. The method also involves receiving a message advertising a first sequence number from another node in the cluster, and subsequently using the first sequence number as the current sequence number.
1. A method comprising:
receiving a message advertising a first sequence number from another node in a cluster;
using the first sequence number as a current sequence number;
associating the current sequence number with each of a plurality of write operations included in a set of independent write operations;
selecting a new sequence number, in response to detecting that one of the write operations has been performed, wherein
the write operations are initiated by an application, and
the new sequence number is based on a value of the current sequence number being used at each node in the cluster;
sending a message advertising the new sequence number to each other node in the cluster; and
preventing any of the write operations in the set of independent write operations from returning to the application until the new sequence number has been advertised to each node included in the cluster.
2. The method of claim 1 , wherein the associating the current sequence number with a first write operation in the set of independent write operations comprises:
associating the current sequence number with each block of data being modified by the first write operation.
3. The method of claim 2 , wherein the associating the current sequence number with each block of data being modified by the first write operation comprises one of:
appending the current sequence number to each block of data, prior to transferring each block of data via a network; and
writing the current sequence number to a storage location associated with each block of data.
4. A node comprising:
a processor; and
a memory storing program instructions executable by the processor to implement:
a sequence number module, wherein
the sequence number module is configured to
receive a message advertising a first sequence number from another node in a cluster;
use the first sequence number as a current sequence number;
generate a new sequence number, wherein
the new sequence number is based on a value of a current sequence number being used at each node included in the cluster, and
advertise the new sequence number to each other node included in the cluster; and
an ordered write module configured to:
associate the current sequence number with each of a plurality of write operations included in a set of independent write operations;
request a new sequence number from the sequence number module, in response to detecting that one of the write operations in the set of independent write operations has been performed, wherein
the write operations are initiated by an application; and
prevent any of the write operations in the set of independent write operations from returning to the application until the new sequence number has been advertised to each node included in the cluster.
5. A system comprising:
a plurality of nodes coupled by a network, wherein
each of the nodes is configured to receive write requests and initiate corresponding write operations;
each of the nodes is configured to receive a message advertising a first sequence number from another node in a cluster;
each of the nodes is configured to use the first sequence number as a current sequence number;
each of the nodes is configured to associate the current sequence number with each of a plurality of write operations included in a set of independent write operations;
each of the nodes is configured to generate and advertise a new sequence number to each other node, in response to detecting that any write operation within the set of independent write operations has been performed, wherein
the set of independent write operations is initiated by an application, and
the new sequence number is based on a value of the current sequence number being used at each node in the cluster; and
each of the nodes is configured to prevent any of the write operations in the set of independent write operations from returning to the application until the new sequence number has been advertised to each node included in the cluster.
6. A computer readable storage medium comprising program instructions executable to:
detect reception of a message advertising a first sequence number from another node in a cluster;
use the first sequence number as a current sequence number;
associate the current sequence number with each of a plurality of write operations included in a set of independent write operations;
select a new sequence number, in response to detecting that one of the write operations has been performed, wherein
the write operations are initiated by an application, and
the new sequence number is based on a value of the current sequence number being used at each node in the cluster;
send a message advertising the new sequence number to each other node in the cluster; and
prevent any of the write operations in the set of independent write operations from returning to the application until the new sequence number has been advertised to each node included in the cluster.
7. A system comprising:
means for detecting reception of a message advertising a first sequence number from another node in a cluster;
means for setting the first sequence number as a current sequence number;
means for associating the current sequence number with each of a plurality of write operations included in a set of independent write operations;
means for selecting a new sequence number, in response to detecting that one of the write operations has been performed, wherein
the write operations are initiated by an application, and
the new sequence number is based on a value of the current sequence number being used at each node in the cluster;
network interface means for sending a message advertising the new sequence number to each other node in the cluster; and
means for preventing any of the write operations in the set of independent write operations from returning to the application until the new sequence number has been advertised to each node included in the cluster.