Hybrid TS7700 Grid configurations

Edit online

A hybrid TS7700 Grid is one that contains an intermix of one or more TS7700 tape-attach Clusters and one or more TS7700 disk-only Clusters that are configured together within the same grid.

The hybrid grid configuration is used for archival and data retention purposes where data consistency shifts from disk-only clusters (clusters that do not connect to a physical library) to clusters that do connect to a physical library. Data that is contained within any TS7700 Cluster can be accessed by any peer cluster, regardless of either cluster's connection to a physical library.

A hybrid TS7700 Grid can be configured by using four, three, or twobetween two and six TS7700 Clusters.

Seven- or eight- cluster hybrid

Seven- and eight-cluster hybrid grid configurations are supported by RPQ only. Possible configurations of seven and eight cluster grids include:

Three production sites each with two clusters, plus a one- or two-cluster disaster recovery site: Each production site contains one disk-only cluster combined with one cluster attached to a physical library. When part of a seven-cluster grid, the disaster recovery site contains one cluster attached to a physical library. When part of a eight-cluster grid, the disaster recovery site also contains a disk-only cluster.; Each production site writes and reads only to its local clusters and all data that is written within either production site is replicated to the adjacent cluster and all disaster recovery clusters.
One production site with four clusters, plus one disaster recovery site with four clusters: Each site contains two disk-only clusters and one cluster attached to a physical library. This configuration is available only with a eight-cluster grid.; Data that is written to a production cluster is replicated to one adjacent cluster and to at least two disaster recovery clusters.

Figure 1. Example of a TS7760 and TS7770 8-way hybrid Grid configuration

Five- or six- cluster hybrid

Possible configurations of five and six cluster grids include:

Two production sites each with two clusters, plus a one- or two-cluster disaster recovery site: Each production site contains one disk-only cluster combined with one cluster attached to a physical library. When part of a five-cluster grid, the disaster recovery site contains one cluster attached to a physical library. When part of a six-cluster grid, the disaster recovery site also contains a disk-only cluster.; Each production site writes and reads only to its local clusters and all data that is written within either production site is replicated to the adjacent cluster and all disaster recovery clusters.
One production site with three clusters, plus one disaster recovery site with three clusters: Each site contains two disk-only clusters and one cluster attached to a physical library. This configuration is available only with a six-cluster grid.; Data that is written to a production cluster is replicated to one adjacent cluster and to at least two disaster recovery clusters.

Four-cluster hybrid

There are two common configurations of a four-cluster grid:

Two disk-only production clusters, plus two remote clusters attached to physical libraries: Disk-only clusters are used for production data, while clusters attached to physical libraries are used for archival or disaster recovery purposes. Optimally, the clusters that are attached to physical libraries are at metro distances to separate long-term data from a single point of loss.; All clusters are configured for high availability. However, the production clusters are also configured for high cache hit rates. Long-term retention data migrates to the clusters attached to physical libraries. The copy export function can be used on these clusters to produce a second or third offsite copy of long-term retention data.
Identical production and remote configurations: One configuration is duplicated at multiple sites: one disk-only cluster combined with one cluster attached to a physical library. This provides a high-availability solution for both production and archival data centers, while distancing the clusters attached to physical libraries for disaster recovery purposes. All clusters are configured for high cache hit rates.

Three-cluster hybrid

One disk-only cluster is used for production, as is one of the clusters attached to a physical library. The other cluster that is attached to a physical library is used for archival or disaster recovery purposes. This solution is a high availability production configuration that also provides the benefits of high cache hit rates for recent data.

Two-cluster hybrid

One disk-only cluster is used for production, while one cluster attached to a physical library is used for archival or disaster recovery purposes. Copy export can be used to produce a second offsite copy of the long-term data that resides only in the cluster attached to a physical library.

Figure 2. Example of a TS7760 hybrid configuration

Table 1 summarizes the possible configurations of a hybrid TS7700 Grid.

Table 1. Hybrid grid configurations
Total clusters	Site A	Site B	Site C	Expectations
8	Production 1 disk-only cluster 1 library-attached cluster	Production 1 disk-only cluster 1 library-attached cluster	Backup 1 disk-only cluster 1 library-attached cluster	High availability at two sites. Remote access for volumes no longer in disk-only cache. Optimal cache hits of newer content at production sites. Copy export is optionally used to create offsite copies of long-term content.
8	Production and backup 2 disk-only clusters 1 library-attached cluster	Production and backup 2 disk-only clusters 1 library-attached cluster	No third site	High availability at both sites. Local access for volumes no longer in disk-only cache. Copy export is optionally used to create offsite copies of long-term content.
7	Production 1 disk-only cluster 1 library-attached cluster	Production 1 disk-only cluster 1 library-attached cluster	Backup 1 1 library-attached cluster	High availability at both sites. Local access for volumes no longer in disk-only cache. Optimal cache hits of newer content at production sites. Copy export is optionally used to create offsite copies of long-term content.
6	Production 1 disk-only cluster 1 library-attached cluster	Production 1 disk-only cluster 1 library-attached cluster	Backup 1 disk-only cluster 1 library-attached cluster	High availability at two sites. Remote access for volumes no longer in disk-only cache. Optimal cache hits of newer content at production sites. Copy export is optionally used to create offsite copies of long-term content.
6	Production and backup 2 disk-only clusters 1 library-attached cluster	Production and backup 2 disk-only clusters 1 library-attached cluster	No third site	High availability at both sites. Local access for volumes no longer in disk-only cache. Copy export is optionally used to create offsite copies of long-term content.
5	Production 1 disk-only cluster 1 library-attached cluster	Production 1 disk-only cluster 1 library-attached cluster	Backup 1 library-attached cluster	High availability at both sites. Local access for volumes no longer in disk-only cache. Optimal cache hits of newer content at production sites. Copy export is optionally used to create offsite copies of long-term content.
4	Production 2 disk-only clusters	Backup 2 library-attached clusters	No third site	High availability at both sites. Remote access for volumes no longer in disk-only cache. Optimal cache hits of newer content at the production site. Copy export is optionally used to create offsite copies of long-term content.
4	Production 1 disk-only cluster 1 library-attached cluster	Production or backup 1 disk-only cluster 1 library-attached cluster		High availability at both sites. Local access to volumes on tape. Second production site as backup. Optimal cache hits at both production and backup sites.
3	Production 1 disk-only cluster 1 library-attached cluster	Backup 1 library-attached cluster		High availability at the production site. Can require remote mount if local library-attached cluster is down and volume is not in cache. Optimal cache hits at the production site.
2	Production 1 disk-only cluster 1 library-attached cluster	No remote clusters		High availability for recently used volumes. Copy export is optionally used to create offsite copies of long-term content. Loss of cluster attached to physical library can prevent access to older volumes. For scheduled outages, a pre-removal of volumes from the disk-only cluster would be required to assure enough cache space is available.

Important: Copy policies must be set up correctly across the hybrid grid. Optimally, all volumes that are created on a disk-only cluster have a copy policy that creates a copy on one or more clusters attached to a physical library. Otherwise, the automatic removal cannot remove volumes from the disk-only cache, leading to the disk-only cache reaching its maximum capacity. If this happens, a manual intervention is required to reduce the volume use of the disk-only cluster.

TS7700 Clusters that attach to a physical library have a much higher total storage capacity than disk-only TS7700 Clusters. In a hybrid grid, a copy export from a cluster that is attached to a physical library can export all the contents of the library including any hybrid-migrated data that resides only on the TS7700 Clusters attached to a physical library. Therefore, when a complete failure occurs to a copy export parent site, it is possible that the copy exported volumes are the only source for the hybrid-migrated data. A recovery in this scenario can occur by initiating a disaster recovery process into a new TS7700 Cluster attached to a physical library by using the copy exported volumes. This allows the data to be merged back into the composite library and any preexisting TS7700 Clusters remain present with valid content. This disaster recovery approach, as opposed to a copy export recovery approach, does not require the recovery point to be a stand-alone TS7700 Cluster attached to a physical library.

Automatic removal policy

The automatic removal policy frees up available cache space on disk-only TS7700 Clusters in a hybrid grid. The Grid Summary page of the TS7700 Management Interface Tape Volume Cache panel can be used to enable a temporary removal threshold:

(Permanent) Removal Thresholds

The default, or permanent, Removal Threshold is used to prevent a cache overrun condition in a TS7700 Cluster that is configured as part of a grid. By default it is a 4 TB value (3 TB fixed plus 1 TB) that, when taken with the amount of used cache, defines the upper size limit for a TS7700 Cache or for a TS7700 Tape Attach CP0. Above this threshold, virtual volumes begin to be removed from a TS7700 Cache. Virtual volumes are removed from a TS7700 Cache in this order:

Volumes in scratch categories
Private volumes least recently used, using the enhanced removal policy definitions

Once removal begins, the TS7700 continues to remove virtual volumes until the Stop Threshold is met. The Stop Threshold is a value that is the Removal Threshold minus 500 GB.

A particular virtual volume cannot be removed from a TS7700 Cache until the TS7700 verifies that a consistent copy exists on a peer cluster. If a peer cluster is not available, or a volume copy has not yet completed, the virtual volume is not a candidate for removal until the appropriate number of copies can be verified later.

Note: The default removal threshold can be changed from the command line, by using the following library request command:

LIBRARY REQUEST library-name,SETTING,CACHE,REMVTHR,{VALUE}

Temporary Removal Thresholds

The Temporary Removal Threshold lowers the default Removal Threshold to a value lower than the Stop Threshold in anticipation of a Service mode event.

The Temporary Removal Threshold value must be equal to or greater than the expected amount of compressed host workload written, copied, or both to the TS7700 during the service outage. The Temporary Removal Threshold is 4 TB providing 5 TB (4 TB plus 1 TB) of free space exists, but you can lower the threshold to any value between 2 TB and full capacity minus 2 TB.

All TS7700 Clusters in the grid that remain available automatically lower their Removal Thresholds to the Temporary Removal Threshold value defined for each. Each TS7700 Cluster may use a different Temporary Removal Threshold. The default Temporary Removal Threshold value is 4 TB or an additional 1 TB more data than the default removal threshold of 3 TB. Each TS7700 Cluster will use its defined value until any cluster in the grid enters Service mode or the temporary removal process is canceled. The cluster that initiates the temporary removal process will not lower its own removal threshold during this process.

So that a disaster recovery test can access all production host-written volumes, automatic removal is temporarily disabled while disaster recovery write protect is enabled on a disk-only cluster. When the write protect state is lifted, automatic removal returns to normal operation.