Optimized message retransmission mechanism for distributed storage virtualization directory system

Patent metadata

Field	Value
Publication number	US-9806896-B2
Application number	US-201414558528-A
Country	US
Kind code	B2
Filing date	Dec 2, 2014
Priority date	Sep 30, 2014
Publication date	Oct 31, 2017
Grant date	Oct 31, 2017

Abstract

A method of retransmitting a multicast message through a unicast channel in a virtual distributed storage system comprising a cluster of nodes that includes a master node and a set of agent nodes is provided. Each node maintains a copy of a directory of the virtual distributed storage system. The master node multicasts messages to the agent node through a multicast channel to update the directory. The method determines that a particular message to update the directory exceeds a maximum size limit for transmit to the agent nodes through the multicast channel. The method multicasts a retransmission trigger message from the master node to each agent node. The method receives a retransmission request message at the master node from each of the agent nodes through a unicast channel. The method sends the particular message to update the directory from the master node to each agent node through the unicast channel.

First claim

What is claimed is:

1. A method of retransmitting a multicast message through a unicast channel in a virtual distributed storage system comprising a cluster of nodes comprising a master node and a set of agent nodes, each node in the cluster of nodes maintaining a copy of a directory of the virtual distributed storage system, the master node multicasting messages to the agent nodes through a multicast channel to update the directory, the method comprising:
at the master node, determining that a particular message to update the directory exceeds a maximum size limit for transmission to the agent nodes through the multicast channel, the particular message comprising an identification of the message;
multicasting a retransmission trigger message from the master node to each of the agent nodes, the retransmission trigger message comprising the identification of the particular message, wherein the master and the agent nodes utilize a protocol to communicate the multicast and unicast messages, wherein said protocol does not allow the master node to send an unsolicited unicast message to an agent node;
in response to the retransmission trigger message, receiving a retransmission request message at the master node from each of the agent nodes through a unicast channel established between the master node and the agent node; and
in response to the receiving of the retransmission request message from each agent node, sending the particular message to update the directory from the master node to the agent node through the unicast channel established between the master node and the agent node.

2. The method of claim 1 , wherein the identification of the particular message is a sequence number of the message, wherein each directory update message transmitted from the master node comprises a sequence number that increases for each consecutive directory update message.

3. The method of claim 1 further comprising:
receiving an update to the directory of the distributed storage system at the master node from an agent node; and
updating the copy of directory of the distributed storage system maintained by the master node based on the received update,
wherein the directory update message is sent from the master node to each agent node to update the copy of the directory of the distributed storage system maintained by each agent node based on the received update.

4. The method of claim 1 , wherein each node is a physical computing device executing a plurality of data compute nodes (DCNs), wherein the update to the directory of the distributed storage system is received at the master node from an agent node when a DCN on the agent node updates an object in the distributed storage system.

5. The method of claim 4 , wherein a DCN is a virtual machine (VM).

6. The method of claim 1 , wherein the protocol is an open systems interconnection (OSI) model layer 2 (L2) protocol.

7. The method of claim 1 , wherein each node is a physical computing device comprising a set of local storage resources, wherein the virtual distributed storage system comprises the local storage resources of each node in the cluster of nodes.

8. The method of claim 1 , wherein the virtual distributed storage system comprises a set of storage objects, each storage object in the set of storage objects maintained by a node in the cluster of nodes.

9. A non-transitory machine readable medium storing a program for retransmitting a multicast message through a unicast channel in a virtual distributed storage system comprising a cluster of nodes comprising a master node and a set of agent nodes, each node in the cluster of nodes maintaining a copy of a directory of the virtual distributed storage system, the master node multicasting messages to the agent nodes through a multicast channel to update the directory, the program executable by a set of processing units of the master node, the program comprising sets of instructions for:
determining that a particular message to update the directory exceeds a maximum size limit for transmission to the agent nodes through the multicast channel, the particular message comprising an identification of the message;
multicasting a retransmission trigger message from the master node to each of the agent nodes, the retransmission trigger message comprising the identification of the particular message, wherein the master and the agent nodes utilize a protocol to communicate the multicast and unicast messages, wherein said protocol does not allow the master node to send an unsolicited unicast message to an agent node;
receiving, in response to the retransmission trigger message, a retransmission request message at the master node from each of the agent nodes through a unicast channel established between the master node and the agent node; and
sending, in response to the receiving of the retransmission request message from each agent node, the particular message to update the directory from the master node to the agent node through the unicast channel established between the master node and the agent node.

10. The non-transitory machine readable medium of claim 9 , wherein the identification of the particular message is a sequence number of the message, wherein each directory update message transmitted from the master node comprises a sequence number that increases for each consecutive directory update message.

11. The non-transitory machine readable medium of claim 9 , the program further comprising sets of instructions for:
receiving an update to the directory of the distributed storage system at the master node from an agent node; and
updating the copy of directory of the distributed storage system maintained by the master node based on the received update,
wherein the directory update message is sent from the master node to each agent node to update the copy of the directory of the distributed storage system maintained by each agent node based on the received update.

12. The non-transitory machine readable medium of claim 9 , wherein each node is a physical computing device executing a plurality of data compute nodes (DCNs), wherein the update to the directory of the distributed storage system is received at the master node from an agent node when a DCN on the agent node updates an object in the distributed storage system.

13. The non-transitory machine readable medium of claim 12 , wherein a DCN is a virtual machine (VM).

14. The non-transitory machine readable medium of claim 9 , wherein the protocol is an open systems interconnection (OSI) model layer 2 (L2) protocol.

15. The non-transitory machine readable medium of claim 9 , wherein each node is a physical computing device comprising a set of local storage resources, wherein the virtual distributed storage system comprises the local storage resources of each node in the cluster of nodes.

16. The non-transitory machine readable medium of claim 9 , wherein the virtual distributed storage system comprises a set of storage objects, each storage object in the set of storage objects maintained by a node in the cluster of nodes.

Assignees

• Nicira Inc

Inventors

• Xiang Enning
• Delgado Jorge Guerra

Classifications

• H04L1/1685

  the supervisory signal being transmitted in response to a specific request, e.g. to a polling signal · CPC title

• G06F2009/45595

  Network integration; Enabling network access in virtual machine instances · CPC title

• H04L49/552

  by ensuring the integrity of packets received through redundant connections · CPC title

• H04L2001/0093

  Point-to-multipoint · CPC title

• H04L12/1881

  with schedule organisation, e.g. priority, sequence management · CPC title