Site and storage tier aware reference resolution
A site and storage tier aware technique replicates data as one or more recovery points (RPs) from a primary site to a secondary site in a multi-site data replication environment. A storage tier aware reference resolver determines (i) an amount of RP data transfer associated with the replication and (ii) location information associated with a cloud storage tier storing the RP data in an object store. The storage tier aware reference resolution aspect provides two additional factors to consider when retrieving data of a reference RP from cloud storage: (iii) the time (duration) needed to retrieve the data and (iv) the cost (financial expense) needed to retrieve the data. In addition, a site aware reference resolution aspect of the technique determines an optimal RP to use as the reference RP and involves consideration of (v) which RPs have been replicated from the primary site to the secondary site and (vi) which RPs have been retained for storage at the sites.
1 . A method comprising:
replicating and incrementally updating data as one or more recovery points (RPs) from a primary site to one or more storage tiers of a multi-tiered archival storage system at a secondary site, wherein the secondary site retains an inventory of RPs larger than that of the primary site;
initiating by the primary site a protocol message exchange with the secondary site to identify one or more candidate locations to use for retrieval of the data based on locations of the RPs among the one or more storage tiers, wherein the protocol message exchange includes constraints provided by the primary site of an expense to extract the data and a duration to extract the data from the one or more storage tiers; and
retrieving the data at the primary site from the secondary site using incremental differences of the data based on a reference RP, wherein the reference RP is rendered from a reference resolution selection based on the constraints provided by the primary site as well as the larger inventory of RPs retained at the secondary site and the identified candidate location of the data.
2 . The method of claim 1 , further comprising:
providing a list of the candidate locations to the primary site from the secondary site; and
pruning the list of candidate locations, wherein the identification of the candidate location to use for the data retrieval is made at the primary site.
3 . The method of claim 1 , further comprising:
replicating the RP to another secondary site having a retention policy longer than that for the RP at the secondary site, wherein the identified candidate location includes the RP at the another secondary site.
4 . The method of claim 1 , wherein the retrieved data of the RP is from a plurality of candidate locations and the identification of the candidate location to use for retrieval of the data includes an aggregation of the plurality of candidate locations according to the constraints.
5 . The method of claim 1 , wherein the constraints optimize for a lowest duration to retrieve the data.
6 . The method of claim 1 , further comprising cascade replicating the RP to a tertiary site from the secondary site, wherein the replicated RP to the tertiary site is included in the candidate locations.
7 . The method of claim 1 , wherein the primary site includes the RP with a retention policy shorter than that of the secondary site.
8 . The method of claim 1 , wherein the identified candidate location at the archival storage system is stored in an index data structure created when the RP is replicated to the secondary site.
9 . A non-transitory computer readable medium including program instructions for execution on a processor, the program instructions configured to:
replicate and incrementally update data as one or more recovery points (RPs) from a primary site to one or more storage tiers of a multi-tiered archival storage system at a secondary site, wherein the secondary site retains an inventory of RPs larger than that of the primary site;
initiate by the primary site a protocol message exchange with the secondary site to identify one or more candidate locations to use for retrieval of the data based on locations of the RPs among the one or more storage tiers, wherein the protocol message exchange includes constraints provided by the primary site of an expense to extract the data and a duration to extract the data from the one or more storage tiers; and
retrieve the data at the primary site from the secondary site using incremental differences of the data based on a reference RP, wherein the reference RP is rendered from a reference resolution selection based on the constraints provided by the primary site as well as the larger inventory of RPs retained at the secondary site and the identified candidate location of the data.
10 . The non-transitory computer readable medium of claim 9 wherein the program instructions are further configured to:
provide a list of the candidate locations to the primary site from the secondary site; and
prune the list of candidate locations, wherein the identification of the candidate location to use for the data retrieval is made at the primary site.
11 . The non-transitory computer readable medium of claim 9 wherein the program instructions are further configured to:
replicate the RP to another secondary site having a retention policy longer than that for the RP at the secondary site, wherein the identified candidate location includes the RP at the another secondary site.
12 . The non-transitory computer readable medium of claim 9 wherein the retrieved data of the RP is from a plurality of candidate locations and the identification of the candidate location to use for retrieval of the data includes an aggregation of the plurality of candidate locations according to the constraints.
13 . The non-transitory computer readable medium of claim 9 wherein the constraints optimize for a lowest duration to retrieve the data.
14 . The non-transitory computer readable medium of claim 9 wherein the program instructions are further configured to cascade replicate the RP to a tertiary site from the secondary site, wherein the replicated RP to the tertiary site is included in the candidate locations.
15 . The non-transitory computer readable medium of claim 9 wherein the primary site includes the RP with a retention policy shorter than that of the secondary site.
16 . The non-transitory computer readable medium of claim 9 wherein the identified candidate location at the archival storage system is stored in an index data structure created when the RP is replicated to the secondary site.
17 . An apparatus comprising:
a network connecting a node of a primary site to a secondary site, the node having a processor configured to execute program instructions configured to:
replicate and incrementally update data as one or more recovery points (RPs) from a primary site to one or more storage tiers of a multi-tiered archival storage system at a secondary site, wherein the secondary site retains an inventory of RPs larger than that of the primary site;
initiate by the primary site a protocol message exchange with the secondary site to identify one or more candidate locations to use for retrieval of the data based on locations of the RPs among the one or more storage tiers, wherein the protocol message exchange includes constraints provided by the primary site of an expense to extract the data and a duration to extract the data from the one or more storage tiers; and
retrieve the data at the primary site from the secondary site using incremental differences of the data based on a reference RP, wherein the reference RP is rendered from a reference resolution selection based on the constraints provided by the primary site as well as the larger inventory of RPs retained at the secondary site and the identified candidate location of the data.
18 . The apparatus of claim 17 wherein the program instructions are further configured to:
provide a list of the candidate locations to the primary site from the secondary site; and
prune the list of candidate locations, wherein the identification of the candidate location to use for the data retrieval is made at the primary site.
19 . The apparatus of claim 17 wherein the program instructions are further configured to:
replicate the RP to another secondary site having a retention policy longer than that for the RP at the secondary site, wherein the identified candidate location includes the RP at the another secondary site.
20 . The apparatus of claim 17 wherein the retrieved data of the RP is from a plurality of candidate locations and the identification of the candidate location to use for retrieval of the data includes an aggregation of the plurality of candidate locations according to the constraints.
21 . The apparatus of claim 17 wherein the constraints optimize for a lowest duration to retrieve the data.
22 . The apparatus of claim 17 wherein the program instructions are further configured to cascade replicate the RP to a tertiary site from the secondary site, wherein the replicated RP to the tertiary site is included in the candidate locations.
23 . The apparatus of claim 17 wherein the primary site includes the RP with a retention policy shorter than that of the secondary site.
24 . The apparatus of claim 17 wherein the identified candidate location at the archival storage system is stored in an index data structure created when the RP is replicated to the secondary site.