Methods and systems for maintaining cache coherency between nodes in a clustered environment by performing a bitmap lookup in response to a read request from one of the nodes
Disclosed herein are methods, systems, and processes to provide coherency across disjoint caches in clustered environments. It is determined whether a data object is owned by an owner node, where the owner node is one of multiple nodes of a cluster. If the owner node for the data object is identified by the determining, a request is sent to the owner node for the data object. However, if the owner node for the data object is not identified by the determining, selects a node in the cluster is selected as the owner node, and the request for the data object is sent to the owner node.
1. A computer-implemented method implemented in a computer system, comprising:
maintaining lock information for a plurality of data objects, wherein
the lock information is maintained by a master node,
the master node is one of a plurality of nodes of a cluster, and
the lock information identifies each one of a plurality of owner nodes holding a lock on each data object of the plurality of data objects;
receiving a query from a reader node, wherein
the reader node is a node of the plurality of nodes other than the master node,
each owner node of the plurality of owner nodes is a node of the plurality of nodes other than the master node or the reader node, and
the query identifies a data object;
determining whether the data object is among the plurality of data objects; and
in response to a determination that the data object is among the plurality of data objects,
attempting to identify an owner node holding the lock on the data object, wherein the owner node is a one of the plurality of owner nodes, and
in response to the owner node being identified, sending a request for the data object to the owner node.
2. The computer-implemented method of claim 1 , further comprising:
receiving the request at the owner node, wherein
the request is for an updated copy of the data object stored in a local storage device of the owner node;
acquiring a lock for the data object, wherein
the lock is granted by a global lock manager that implements a distributed locking mechanism in the cluster; and
sending the updated copy of the data object to the reader node.
3. The computer-implemented method of claim 1 , wherein
the lock information is implemented using a bitmap,
each bit in the bitmap represents a corresponding data object of the plurality of data objects, and
the method further comprises
performing a bitmap lookup on the bitmap,
determining an availability of a data object, wherein
the availability of the data object is determined based, at least in part, on a result of the bitmap lookup, and
in response to the data object being available, permitting the reader node to access the data object.
4. The computer-implemented method of claim 3 , wherein
the lock information is maintained by a plurality of global lock managers executed by the master node, and
the plurality of global lock managers maintain owner node information coherently by maintaining a consistent copy of the lock information at each node of the plurality of nodes.
5. The computer-implemented method of claim 1 , further comprising:
maintaining owner node information, wherein
the lock information is maintained by a plurality of global lock managers executed by the master node, and
the plurality of global lock managers maintain the owner node information coherently.
6. The computer-implemented method of claim 5 , the attempting comprises:
accessing the owner node information, wherein
the owner node information is accessed in a cluster-wide manner by the plurality of global lock managers.
7. The computer-implemented method of claim 5 , wherein
the plurality of global lock managers implement a distributed locking mechanism that provides metadata coherency and cache coherency across the plurality of nodes.
8. The computer-implemented method of claim 1 , wherein
the lock information is maintained at the master node, and
the data object is stored in a shared storage device of the cluster.
9. The computer-implemented method of claim 1 , further comprising:
maintaining the lock information, wherein
the lock information is maintained by a global lock manager,
the lock is a lock type that is specific to maintenance of cluster-wide ownership of data objects.
10. The computer-implemented method of claim 9 , further comprising:
the lock type minimizes memory footprint across the cluster,
the lock type facilitates determination of owner node identity, and
the lock type permits storage of owner node information indicating multiple owner nodes, when the data object is cached at multiple locations.
11. The computer-implemented method of claim 1 , further comprising:
attempting to acquire the lock for the data object, wherein
the lock is managed by a global lock manager that implements a distributed locking mechanism in the cluster.
12. The computer-implemented method of claim 11 , further comprising:
in response to the lock being acquired,
permitting the reader node to access the data object, comprising
receiving a query for the data object, wherein
the query is received from the reader node, and
in response to a result of the query indicating that the data object is available,
receiving a request to read the data object, and
in response to the request to read the data object, providing the reader node with read access to the data object.
13. The computer-implemented method of claim 11 , further comprising:
in response to the lock not being acquired,
selecting another node of the plurality of nodes, wherein
the selecting is performed by the reader node, and
the another node is one of the plurality of nodes other than then node.
14. The computer-implemented method of claim 11 , further comprising:
granting the lock, wherein
the lock is granted by the global lock manager,
the lock comprises lock information regarding the data object, and
the lock information is maintained by the global lock manager.
15. A non-transitory computer-readable storage medium, comprising program instructions, which, when executed by one or more processors of a computing system, perform a method comprising:
maintaining lock information for a plurality of data objects, wherein
the lock information is maintained by a master node,
the master node is one of a plurality of nodes of a cluster, and
the lock information identifies each one of a plurality of owner nodes holding a lock on each data object of the plurality of data objects;
receiving a query from a reader node, wherein
the reader node is a node of the plurality of nodes other than the master node,
each owner node of the plurality of owner nodes is a node of the plurality of nodes other than the master node or the reader node, and
the query identifies a data object;
determining whether the data object is among the plurality of data objects; and
in response to a determination that the data object is among the plurality of data objects,
attempting to identify an owner node holding the lock on the data object, wherein the owner node is a one of the plurality of owner nodes, and
in response to the owner node being identified, sending a request for the data object to the owner node.
16. The non-transitory computer-readable storage medium of claim 15 , wherein the method further comprises:
maintaining owner node information, wherein
the lock information is maintained by a plurality of global lock managers executed by the master node, and
the plurality of global lock managers maintain the owner node information coherently.
17. The non-transitory computer-readable storage medium of claim 16 , wherein the method further comprises:
accessing the owner node information, wherein
the owner node information is accessed in a cluster-wide manner by the plurality of global lock managers.
18. The non-transitory computer-readable storage medium of claim 16 , wherein the method further comprises:
the plurality of global lock managers implement a distributed locking mechanism that provides metadata coherency and cache coherency across the plurality of nodes.
19. The non-transitory computer-readable storage medium of claim 15 , wherein the method further comprises:
attempting to acquire the lock for the data object, wherein
the lock is managed by a global lock manager that implements a distributed locking mechanism in the cluster; and
granting the lock, wherein
the lock is granted by the global lock manager,
the lock comprises lock information regarding the data object, and
the lock information is maintained by the global lock manager.
20. A system comprising:
one or more processors; and
a memory coupled to the one or more processors, wherein the memory stores program instructions executable by the one or more processors to perform a method comprising
maintaining lock information for a plurality of data objects, wherein
the lock information is maintained by a master node,
the master node is one of a plurality of nodes of a cluster, and
the lock information identifies each one of a plurality of owner nodes holding a lock on each data object of the plurality of data objects,
receiving a query from a reader node, wherein
the reader node is a node of the plurality of nodes other than the master node,
each owner node of the plurality of owner nodes is a node of the plurality of nodes other than the master node or the reader node, and
the query identifies a data object,
determining whether the data object is among the plurality of data objects, and
in response to a determination that the data object is among the plurality of data objects,
attempting to identify an owner node holding the lock on the data object, wherein
the owner node is a one of the plurality of owner nodes, and
in response to the owner node being identified, sending a request for the data object to the owner node.