Performance of replication system with small number of IP links available
A method is provided for use in a storage system, comprising: identifying a first process that is arranged to execute a first type-1 node and a first type-2 node of the storage system, the first type-1 node being assigned a communication link for transmitting replication data to a target system, the first type-2 node being arranged to execute I/O requests associated with a first set of addresses in an address space; identifying a second process that is arranged to execute a second type-1 node and a second type-2 node of the storage system, the second type-1 node being not being assigned any communication link for transmitting replication data to a target system, the second type-2 node being arranged to execute I/O requests associated with a second set of addresses in the address space; and transferring at least one of the addresses in the first set to the second set.
1. A method for use in a storage system, comprising:
identifying a first process that is arranged to execute a first type-1 node and a first type-2 node of the storage system, the first type-1 node being assigned a communication link for transmitting replication data to a target system, the first type-2 node being arranged to execute I/O requests associated with a first set of addresses in an address space, and the first type-2 node being configured to transmit data replication requests to any of one or more type-1 nodes in the storage system that are assigned communication links for transmitting data to a replication system;
identifying a second process that is arranged to execute a second type-1 node and a second type-2 node of the storage system, the second type-1 node being not being assigned any communication link for transmitting replication data to a target system, the second type-2 node being arranged to execute I/O requests associated with a second set of addresses in the address space, and the second type-2 node being configured to transmit data replication requests to any of the one or more type-1 nodes in the storage system that are assigned communication links for transmitting data to the replication system; and
transferring at least one of the addresses in the first set to the second set, wherein transferring at least one of the addresses in the first set to the second set at least partially corrects for an imbalance between a first workload on the first process and a second workload on the second process, the imbalance resulting from the first type-1 node being the only one of the first type-1 node and the second type-1 node that is assigned a communication link for transmitting data to the replication system.
2. The method of claim 1 , wherein the first type-1 node and the first type-2 node include respective threads that are instantiated within the first processes, and the second type-1 node and the second type-2 node include respective threads that are instantiated within the second process.
3. The method of claim 1 , wherein:
the first type-1 node is configured to: (i) provide to the first type-2 node any incoming I/O requests that are associated with an address from the first set; and (ii) provide to the second type-2 node any incoming I/O requests that are associated with an address from the second set, and
the second type-1 node is configured to: (i) provide to the first type-2 node any incoming I/O requests that are associated with an address from the first set; and (ii) provide to the second type-2 node any incoming I/O requests that are associated with an address from the second set.
4. The method of claim 1 , wherein the first type-1 node includes a first R-node node, the first type-2 node includes a first C-node, the second type-1 node includes a second R-node, and the second type-2 node includes a second C-node.
5. The method of claim 1 , wherein the first process includes a first operating system process and the second process includes a second operating system process.
6. The method of claim 1 , further comprising, storing in a memory, a data structure that: (i) maps a first definition of the first set to the first type-2 node and (ii) maps a second definition of the second set to the second type-2 node, wherein transferring the at least one of the addresses in the first set to the second set includes removing the address from the first definition and adding the address to the second definition.
7. The method of claim 6 , wherein the data structure includes an address-to-C-node (A2C) table.
8. An apparatus, comprising:
a memory; and
at least one processor operatively coupled to the memory, the at least one processor being configured to perform the operations of:
identifying a first process that is arranged to execute a first type-1 node and a first type-2 node of the storage system, the first type-1 node being assigned a communication link for transmitting replication data to a target system, the first type-2 node being arranged to execute I/O requests associated with a first set of addresses in an address space, and the first type-2 node being configured to transmit data replication requests to any of one or more type-1 nodes in the storage system that are assigned communication links for transmitting data to a replication system;
identifying a second process that is arranged to execute a second type-1 node and a second type-2 node of the storage system, the second type-1 node being not being assigned any communication link for transmitting replication data to a target system, the second type-2 node being arranged to execute I/O requests associated with a second set of addresses in the address space, and the second type-2 node being configured to transmit data replication requests to any of the one or more type-1 nodes in the storage system that are assigned communication links for transmitting data to the replication system; and
transferring at least one of the addresses in the first set to the second set, wherein transferring at least one of the addresses in the first set to the second set at least partially corrects for an imbalance between a first workload on the first process and a second workload on the second process, the imbalance resulting from the first type-1 node being the only one of the first type-1 node and the second type-1 node that is assigned a communication link for transmitting data to the replication system.
9. The apparatus of claim 8 , wherein the first type-1 node and the first type-2 node include respective threads that are instantiated within the first processes, and the second type-1 node and the second type-2 node include respective threads that are instantiated within the second process.
10. The apparatus of claim 8 , wherein:
the first type-1 node is configured to: (i) provide to the first type-2 node any incoming I/O requests that are associated with an address from the first set; and (ii) provide to the second type-2 node any incoming I/O requests that are associated with an address from the second set, and
the second type-1 node is configured to: (i) provide to the first type-2 node any incoming I/O requests that are associated with an address from the first set; and (ii) provide to the second type-2 node any incoming I/O requests that are associated with an address from the second set.
11. The apparatus of claim 8 , wherein the first type-1 node includes a first R-node node, the first type-2 node includes a first C-node, the second type-1 node includes a second R-node, and the second type-2 node includes a second C-node.
12. The apparatus of claim 8 , wherein the first process includes a first operating system process and the second process includes a second operating system process.
13. The apparatus of claim 8 , wherein:
the memory is configured to store a data structure that: (i) maps a first definition of the first set to the first type-2 node and (ii) maps a second definition of the second set to the second type-2 node, and
transferring the at least one of the addresses in the first set to the second set includes removing the address from the first definition and adding the address to the second definition.
14. The apparatus of claim 13 , wherein the data structure includes an address-to-type-2 node (A2C) table.
15. A non-transitory computer-readable medium storing one or more processor-executable instructions, which when executed by at least one processor cause the at least one processor to perform the operations of:
identifying a first process that is arranged to execute a first type-1 node and a first type-2 node of the storage system, the first type-1 node being assigned a communication link for transmitting replication data to a target system, the first type-2 node being arranged to execute I/O requests associated with a first set of addresses in an address space, and the first type-2 node being configured to transmit data replication requests to any of one or more type-1 nodes in the storage system that are assigned communication links for transmitting data to a replication system;
identifying a second process that is arranged to execute a second type-1 node and a second type-2 node of the storage system, the second type-1 node being not being assigned any communication link for transmitting replication data to a target system, the second type-2 node being arranged to execute I/O requests associated with a second set of addresses in the address space, and the second type-2 node being configured to transmit data replication requests to any of the one or more type-1 nodes in the storage system that are assigned communication links for transmitting data to the replication system; and
transferring at least one of the addresses in the first set to the second set, wherein transferring at least one of the addresses in the first set to the second set at least partially corrects for an imbalance between a first workload on the first process and a second workload on the second process, the imbalance resulting from the first type-1 node being the only one of the first type-1 node and the second type-1 node that is assigned a communication link for transmitting data to the replication system.
16. The non-transitory computer-readable medium of claim 15 , wherein the first type-1 node and the first type-2 node include respective threads that are instantiated within the first processes, and the second type-1 node and the second type-2 node include respective threads that are instantiated within the second process.
17. The non-transitory computer-readable medium of claim 15 , wherein:
the first type-1 node is configured to: (i) provide to the first type-2 node any incoming I/O requests that are associated with an address from the first set; and (ii) provide to the second type-2 node any incoming I/O requests that are associated with an address from the second set, and
the second type-1 node is configured to: (i) provide to the first type-2 node any incoming I/O requests that are associated with an address from the first set; and (ii) provide to the second type-2 node any incoming I/O requests that are associated with an address from the second set.
18. The non-transitory computer-readable medium of claim 15 , wherein the first type-1 node includes a first R-node node, the first type-2 node includes a first C-node, the second type-1 node includes a second R-node, and the second type-2 node includes a second C-node.
19. The non-transitory computer-readable medium of claim 15 , wherein the first process includes a first operating system process and the second process includes a second operating system process.
20. The non-transitory computer-readable medium of claim 15 , further comprising, storing in a memory, a data structure that: (i) maps a first definition of the first set to the first type-2 node and (ii) maps a second definition of the second set to the second type-2 node, wherein transferring the at least one of the addresses in the first set to the second set includes removing the address from the first definition and adding the address to the second definition.