Workload considerations

Edit online

This topic highlights some of the considerations to bear in mind when you decide what workload to place in the TS7785.

With the TS7785, you can backup and restore existing workloads without requiring application changes. Multiple virtual libraries can be configured to support concurrent workloads. By emulating a TS4500 tape library environment with LTO-5 drives, the TS7785 integrates into your environment in the same way as a physical tape library. For active data, backup and restore operations run from disk cache instead of waiting for physical tape operations.

Throughput

Throughput is the rate at which data moves from one TS7785 cluster to another target, such as another cluster in a grid, a cloud/object target. Throughput is typically measured in MBps. Relevant throughput metrics include host reads, host writes, cloud reads, cloud writes, and cluster-to-cluster transfers.

Drive concurrency

The design of the TS7785 cluster allows transparent access to multiple virtual tape drives on the same virtual library.

Cartridge capacity utilization

One of the key benefits of the TS7785 is its ability to fully use the capacity of the virtual tape cartridges independent of the data set sizes that are written. Another is the ability to manage that capacity effectively without host or user involvement. A virtual cartridge can contain up to 500 GB of data.

Volume caching

Often, one step of a job writes a tape volume and a subsequent step (or job) reads it. The TS7785 improves the efficiency of this process: As data is cached in the TS7785 Cache the rewind time, the robotics time, and load or thread times for the mount are effectively removed. When a job attempts to read a volume that is not in the TS7785 Cache, the virtual cartridge is recalled from a stacked physical volume back into the cache. When a recall is necessary, the time to access the data is greater than if the data were already in the cache. The size of the cache and the use of cache management policies can reduce the number of recalls. Too much recall activity can negatively affect overall throughput of the TS7785.

Fast mount times

When a program issues a fast mount to write data, the TS7785 completes the mount request without having to recall the virtual cartridge into the cache. For workloads that create many tapes, this significantly reduces volume processing overhead times and improves batch window efficiencies.

Fast mount times are further reduced when the optimal fast allocation assistance function is enabled. This function designates one or more clusters as preferred candidates for fast mounts.

Disaster recovery

The grid configuration of the TS7785 is a perfect integrated solution for your disaster recovery data. Multiple TS7785 Clusters can be separated over long distances and interconnected by using an IP infrastructure to provide for automatic data replication. Data that is written to a local TS7785 is accessible at the remote TS7785 as if it was created there. Flexible replication policies make it easy to tailor the replication of data to your business needs.

For more information about disaster recovery configuration, see Configuring for disaster recovery.

Multifile volumes

If your existing workloads use multifile volumes to improve capacity use, those workloads can continue to run on the TS7785 without application changes.

In many cases, manual stacking of multiple files onto a volume is no longer necessary because the TS7785 manages virtual capacity automatically.

Cloud tiering

The TS7785 supports cloud tiering to IBM Storage Deep Archive. You can use cloud tiering to retain less-active data at lower cost while keeping recent data available in cache.

Automatic offload to IBM Storage Deep Archive can help reduce cost per GB for long-term retention workloads.

Management and automation

The TS7785 provides single-pane-of-glass management across all clusters in a grid. This centralized view can simplify monitoring and workload management across multiple sites.

A full-featured REST API supports automation of common management tasks.

Grid network load balancing

For a TS7785 Grid link, the dynamic load balancing function calculates and stores the following information:

  • Instantaneous throughput
  • Number of bytes queued to transfer
  • Total number of jobs queued on both links
  • Whether deferred copy throttling is enabled on the remote node
  • Whether a new job will be throttled (is deferred or immediate)

As a new task starts, a link selection algorithm uses the stored information to identity the link that will most quickly complete the data transfer. The dynamic load balancing function also uses the instantaneous throughput information to identify degraded link performance.

Workload fit

The TS7785 is a good fit for workloads that require fast backup and restore for recent data, low-cost retention for older data, centralized management across multiple clusters, and replication for high availability and disaster recovery.