Method and system for implementing a cluster-wide communication over a single port

1. A system for implementing a handshake between a source node cluster having file domains and a destination node cluster to which said file domains are replicated, the system comprising:

a source node cluster having a plurality of nodes and a replication manager; and

a destination node cluster having a plurality of nodes, a replication manager and a single port manager for each node of the destination node cluster, wherein a number of the nodes at the destination node cluster and the number of nodes at the source node cluster is unequal,

wherein each of the single port managers is configured to inform the replication manager at the source node cluster of the node at the destination cluster node where the replication manager of the destination node cluster is located, via a single port opened in a firewall monitoring communication going between the source node cluster and the destination node cluster, wherein said single port is an only port that is opened for communication via said firewall between said source node cluster and said destination node cluster,

wherein the replication managers of the source and destination node clusters are configured to replicate all files and processes on the nodes of the source node cluster to the nodes of the destination node cluster via said single port,

wherein all replicated file domains register with the single port manager of the node on which they are replicated, and

wherein the single port manager is configured to communicate with the source node cluster via said single port to provide descriptors of replicated file domains, in response to inquiries from the source node cluster.

2. The system according to claim 1 , further comprising a cluster-wide distributed database configured to store the file domain descriptors obtained by each of the single port managers, wherein each of the single port managers is configured to obtain the file domain descriptors of all replicated file domains.

3. The system according to claim 1 , wherein the registration of said file domains register with said single port manager is carried out via a kernel of the respective node.

4. The system according to claim 1 , wherein the source node cluster and a destination node cluster have a different number of nodes.

5. The system according to claim 1 , wherein the source node cluster and a destination node cluster each are part of two distinct file systems.

6. The system according to claim 1 , wherein the source node cluster and a destination node cluster each are part of a same distributed file system.

7. A method for implementing a handshake between a source node cluster having file domains and a destination node cluster to which said file domains are replicated, the method comprising:

providing a source node cluster having a plurality of nodes and a replication manager;

providing a destination node cluster having a plurality of nodes, a replication manager and a single port manager for each node of the destination node cluster, wherein a number of the nodes at the destination node cluster and the number of nodes at the source node cluster is unequal;

informing the replication manager at the source node cluster of the node at the destination cluster node where the replication manager of the destination node cluster is located, via a single port opened in a firewall monitoring communication going between the source node cluster and the destination node cluster, wherein said single port is an only port that is opened for communication via said firewall between said source node cluster and said destination node cluster;

replicating all files and processes on the nodes of the source node cluster to the nodes of the destination node cluster via said single port;

registering all of the replicated file domains with the single port manager of the node on which they are replicated; and

communicating with the source node cluster via said single port to provide descriptors of replicated file domains, in response to inquiries from the source node cluster.

8. The method according to claim 7 , further comprising storing the file domain descriptors obtained by each of the single port managers, wherein each of the single port managers is configured to obtain the file domain descriptors of all replicated file domains.

9. The method according to claim 7 , wherein the registration of said file domains register with said single port manager is carried out via a kernel of the respective node.

10. The method according to claim 7 , wherein the source node cluster and a destination node cluster have a different number of nodes.

11. The method according to claim 7 , wherein the source node cluster and a destination node cluster each are part of two distinct file systems.

12. The method according to claim 7 , wherein the source node cluster and a destination node cluster each are part of a same distributed file system.

13. A non-transitory computer readable medium that includes a set of instructions, that when executed, cause the least one computer processor to:

manage a source node cluster having a plurality of nodes and a replication manager;

manage a destination node cluster having a plurality of nodes, a replication manager and a single port manager for each node of the destination node cluster, wherein a number of the nodes at the destination node cluster and the number of nodes at the source node cluster is unequal;

inform the replication manager at the source node cluster of the node at the destination cluster node where the replication manager of the destination node cluster is located, via a single port opened in a firewall monitoring communication going between the source node cluster and the destination node cluster, wherein said single port is an only port that is opened for communication via said firewall between said source node cluster and said destination node cluster;

replicate all files and processes on the nodes of the source node cluster to the nodes of the destination node cluster via said single port;

registering all of the replicated file domains with the single port manager of the node on which they are replicated; and

communicate with the source node cluster via said single port to provide descriptors of replicated file domains, in response to inquiries from the source node cluster.

14. The non-transitory computer readable medium according to claim 13 , further comprising a set of instructions, that when executed, cause the least one computer processor to store the file domain descriptors obtained by each of the single port managers, wherein each of the single port managers is configured to obtain the file domain descriptors of all replicated file domains.

15. The non-transitory computer readable medium according to claim 13 , wherein the registration of said file domains register with said single port manager is carried out via a kernel of the respective node.

16. The non-transitory computer readable medium according to claim 13 , wherein the source node cluster and a destination node cluster have a different number of nodes.

17. The non-transitory computer readable medium according to claim 13 , wherein the source node cluster and a destination node cluster each are part of two distinct file systems.

18. The non-transitory computer readable medium according to claim 13 , wherein the source node cluster and a destination node cluster each are part of a same distributed file system.