Direct connections to a plurality of storage object replicas in a computer network

• US 8,060,619 B1
• Filed: 11/07/2003
• Issued: 11/15/2011
• Est. Priority Date: 11/07/2003
• Status: Active Grant

• Current Assignee: Veritas Technologies, LLC (Whitehouse Group Ltd.)
• Original Assignee: Symantec Operating Corporation (NortonLifeLock Inc.)
• Inventors: Tewari, Ruchir, Slaughter, Gregory L., Saulpaugh, Thomas E.
• Primary Examiner(s): Vaughn, Jr.; William
• Assistant Examiner(s): Cheema; Umar
• Application Number:
  US10/704,299
• Time in Patent Office: 2,930 Days
• Field of Search: 709/227, 709/229, 709/246, 310/120
• US Class Current: 709/227
• CPC Class Codes: G06F 16/1834 implemented based on peer-to-peer networks, e.g. gnutella p2p communication protocols H04L67/104
• CPC Class Codes: G06F 16/1837 Management specially adapted to peer-to-peer storage networks topology management mechanisms of peer-to-peer networks H04L67/1042
• CPC Class Codes: G06F 16/27 Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
• CPC Class Codes: G06F 16/278 Data partitioning, e.g. horizontal or vertical partitioning
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/104 Peer-to-peer [P2P] networks
• CPC Class Codes: H04L 67/1042 using topology management mechanisms
• CPC Class Codes: H04L 67/1068 Discovery involving direct consultation or announcement among potential requesting and potential source peers
• CPC Class Codes: H04L 67/1078 Resource delivery mechanisms

Abstract

A system and method for efficiently accessing replicas of a storage object. A first node may perform a find operation to determine a plurality of nodes that each store a replica of a first storage object. For each node in the plurality of nodes, the first node may establish a direct connection to the node. The first node may then access the replicas of the first storage object using the respective direct connections to the plurality of nodes. In one embodiment, the nodes may be nodes in a peer-to-peer network. Establishing a direct connection to each storage object replica may enable the first node to communicate with each node in a single hop without having to pass messages via intermediate nodes in the peer-to-peer network.

Claims

1. A non-transitory computer-readable memory medium storing program instructions executable to implement a method comprising:

a first peer node in a peer-to-peer (P2P) network that includes a plurality of peer nodes sending a find message in the P2P network to find other peer nodes that store replicas of a first storage object;

the first peer node receiving one or more reply messages in response to the find message, wherein the one or more reply messages specify addresses of two or more peer nodes that each store a respective replica of the first storage object;

for each respective peer node of the two or more peer nodes;

the first peer node establishing a direct virtual link to the respective peer node using the address of the respective peer node specified by the one or more reply messages; and

the first peer node accessing the respective replica stored on the respective peer node by sending one or more messages directly to the respective peer node through the direct virtual link to the respective peer node without the one or more messages passing through any intermediate peer nodes in the P2P network.

2. The non-transitory computer-readable memory medium of claim 1, wherein the direct virtual links from the first peer node to the two or more peer nodes enable the first peer node to efficiently access each replica of the first storage object without communicating with peer nodes in the P2P network other than the two or more peer nodes on which the respective replicas are stored.

3. The non-transitory computer-readable memory medium of claim 1, wherein, for each respective peer node of the two or more peer nodes, said first peer node accessing the respective replica stored on the respective peer node comprises one or more of:

the first peer node sending the one or more messages to the respective peer node through the direct virtual link to the respective peer node in order to write data to the respective replica stored on the respective peer node; and

/or

the first peer node sending the one or more messages to the respective peer node through the direct virtual link to the respective peer node in order to read data from the respective replica stored on the respective peer node.

4. The non-transitory computer-readable memory medium of claim 1,

wherein each of the two or more peer nodes that store respective replicas of the first storage object hosts a respective instance of a particular address;

wherein said first peer node sending the find message comprises the first peer node sending a find message addressed to the particular address.

5. The non-transitory computer-readable memory medium of claim 1,

wherein the two or more peer nodes that store respective replicas of the first storage object include a second peer node and one or more additional peer nodes, wherein the method implemented by the program instructions further comprises:

the first peer node communicating with a third peer node in order to determine that the second peer node stores a replica of the first storage object;

the first peer node storing routing information indicating how to route messages to the second peer node;

wherein the first peer node sending the find message comprises the first peer node sending the find message to the second peer node using the routing information, wherein the second peer node forwards the find message to each of the one or more additional peer nodes that store replicas of the first storage object.

6. The non-transitory computer-readable memory medium of claim 1,

wherein the plurality of peer nodes in the P2P network are located in a plurality of local area networks (LANs), wherein the first peer node is included in a first LAN of the plurality of LANs;

wherein the find message is sent to peer nodes in multiple LANs;

wherein the two or more peer nodes that store replicas of the first storage object include peer nodes in multiple LANs.

7. The non-transitory computer-readable memory medium of claim 1,

wherein the plurality of peer nodes in the P2P network are located in a plurality of local area networks (LANs), wherein the first peer node is included in a first LAN of the plurality of LANs;

wherein the find message is sent by the first peer node such that the find message is restricted to the first LAN;

wherein each of the two or more peer nodes that store replicas of the first storage object is a node in the first LAN.

8. The non-transitory computer-readable memory medium of claim 1, wherein the method implemented by the program instructions further comprises:

determining a tree data structure comprising a plurality of edges, wherein each edge of the tree is mapped to a respective direct virtual link that connects the first peer node to a respective one of the nodes in the two or more peer nodes;

wherein the first peer node sends the one or more messages to each respective peer node of the two or more peer nodes by sending the one or more messages over the direct virtual links to which the edges of the tree data structure are mapped.

9. The non-transitory computer-readable memory medium of claim 1,

wherein, for each respective peer node of the two or more peer nodes, the direct virtual link from the first peer node to the respective peer node is implemented using a TCP connection.

10. The non-transitory computer-readable memory medium of claim 1,

wherein the first storage object comprises one or more of:

a file;

or

a database.

25. The non-transitory computer-readable memory medium of claim 1,

wherein the first peer node receiving the one or more reply messages in response to the find message comprises the first peer node receiving a single reply message that specifies the addresses of the two or more peer nodes.

11. A system comprising:

a plurality of peer nodes that communicate to form a peer-to-peer (P2P) network;

wherein the plurality of peer nodes execute program instructions to implement a method comprising;

a first peer node of the plurality of peer nodes sending a find message to find other peer nodes that store replicas of a first storage object;

the first peer node receiving one or more reply messages in response to the find message, wherein the one or more reply messages specify addresses of two or more peer nodes that each store a respective replica of the first storage object;

for each respective peer node of the two or more peer nodes;

the first peer node establishing a direct virtual link to the respective peer node using the address of the respective peer node specified by the one or more reply messages; and

the first peer node accessing the respective replica stored on the respective peer node by sending one or more messages directly to the respective peer node through the direct virtual link to the respective peer node without the one or more messages passing through any intermediate peer nodes in the P2P network.

12. The system of claim 11,

wherein the direct virtual links from the first peer node to the two or more peer nodes enable the first peer node to efficiently access each replica of the first storage object without communicating with peer nodes in the P2P network other than the two or more peer nodes on which the respective replicas are stored.

13. The system of claim 11,

wherein, for each respective peer node of the two or more peer nodes, said first peer node accessing the respective replica stored on the respective peer node comprises one or more of:

the first peer node sending the one or more messages to the respective peer node through the direct virtual link to the respective peer node in order to write data to the respective replica stored on the respective peer node; and

/or

the first peer node sending the one or more messages to the respective peer node through the direct virtual link to the respective peer node in order to read data from the respective replica stored on the respective peer node.

14. The system of claim 11,

wherein each of the two or more peer nodes that store respective replicas of the first storage object hosts a respective instance of a particular address;

wherein said first peer node sending the find message comprises the first peer node sending a find message addressed to the particular address.

15. The system of claim 11,

wherein the two or more peer nodes that store respective replicas of the first storage object include a second peer node and one or more additional peer nodes, wherein the method implemented by the program instructions further comprises:

the first peer node communicating with a third peer node in order to determine that the second peer node stores a replica of the first storage object;

the first peer node storing routing information indicating how to route messages to the second peer node;

wherein the first peer node sending the find message comprises the first peer node sending the find message to the second peer node using the routing information, wherein the second peer node forwards the find message to each of the one or more additional peer nodes that store replicas of the first storage object.

16. The system of claim 11,

wherein the plurality of peer nodes in the P2P network are located in a plurality of local area networks (LANs), wherein the first peer node is included in a first LAN of the plurality of LANs;

wherein the find message is sent to peer nodes in multiple LANs;

wherein the two or more peer nodes that store replicas of the first storage object include peer nodes in multiple LANs.

17. The system of claim 11,

wherein the plurality of peer nodes in the P2P network are located in a plurality of local area networks (LANs), wherein the first peer node is included in a first LAN of the plurality of LANs;

wherein the find message is sent by the first peer node such that the find message is restricted to the first LAN;

wherein each of the two or more peer nodes that store replicas of the first storage object is a node in the first LAN.

18. The system of claim 11, wherein the method further comprises:

determining a tree data structure comprising a plurality of edges, wherein each edge of the tree is mapped to a respective direct virtual link that connects the first peer node to a respective one of the nodes in the two or more peer nodes;

wherein the first peer node sends the one or more messages to each respective peer node of the two or more peer nodes by sending the one or more messages over the direct virtual links to which the edges of the tree data structure are mapped.

19. The system of claim 11,

wherein the first storage object comprises one or more of:

a file;

or

a database.

20. A non-transitory computer-readable memory medium storing program instructions executable to implement a method comprising:

a first node in an overlay network that includes a plurality of nodes sending a first message on the overlay network to locate two or more target nodes that store a replica of a first storage object;

the first node receiving address information specifying a respective address for each respective node of the two or more target nodes in response to the first message;

the first node establishing a respective direct virtual link to each respective node of the two or more target nodes using the address information for the target nodes; and

for each respective target node of the two or more target nodes, the first node sending one or more messages directly to the respective target node through the direct virtual link to the respective target node without the one or more messages passing through other nodes in the overlay network.

21. The non-transitory computer-readable memory medium of claim 20,

wherein the nodes in the overlay network communicate to form a peer-to-peer network;

wherein the first node and the two or more target nodes comprise peer nodes in the peer-to-peer network.

22. The non-transitory computer-readable memory medium of claim 20,

wherein said first node receiving address information for the two or more target nodes in response to the first message comprises the first node receiving a single response message in response to the first message, wherein the response message includes address information for all of the target nodes.

23. The non-transitory computer-readable memory medium of claim 20,

wherein each of the two or more target nodes that stores a replica of the first storage object hosts an instance of a first address;

wherein said first node sending the first message to locate the two or more target nodes that store a replica of the first storage object comprises the first node sending the first message to locate all nodes that host an instance of the first address.

24. A method comprising:

a first peer node in a peer-to-peer (P2P) network that includes a plurality of peer nodes sending a find message in the P2P network to find other peer nodes that store replicas of a first storage object;

the first peer node receiving one or more reply messages in response to the find message, wherein the one or more reply messages specify addresses of two or more peer nodes that each store a respective replica of the first storage object;

for each respective peer node of the two or more peer nodes;

the first peer node establishing a direct virtual link to the respective peer node using the address of the respective peer node specified by the one or more reply messages; and

the first peer node accessing the respective replica stored on the respective peer node by sending one or more messages directly to the respective peer node through the direct virtual link to the respective peer node without the one or more messages passing through any intermediate peer nodes in the P2P network.