Skip to main content
  • Place orders quickly and easily
  • View orders and track your shipping status
  • Enjoy members-only rewards and discounts
  • Create and access a list of your products

PowerProtect Data Manager 19.14 Network-Attached Storage User Guide

NAS protection deployment recommendations

Use the sizing information presented here to calculate an optimal deployment for your environment.

PowerProtect Data Manager divides NAS assets into slices by using threshold values. The threshold slice size is 200 GB or a count of one million files, with a tolerance of up to 30%. To avoid creating nonoptimized slices, the tolerance allows PowerProtect Data Manager to deviate from the threshold value and keep data together. For example, a single 260 GB folder is still within the 30% tolerance value and would be considered one slice. Empty placeholder slices help protect the asset layout structure.

Each slice uses one stream. Protection engine parameters provides information about the relationship between streams and NAS protection engines. When calculating the number of NAS protection engines, allow up to 120% to 150% of the estimated slice count for overhead such as placeholder slices.

You can use a sizing tool for NAS to get an approximate value of the number of proxy engines needed to support the production workload. Size recommendations for the number of proxy engines that need to be deployed to protect a certain workload depend on multiple factors, including:

  • Size of the asset
  • Approximate file count and data layout
  • Data change rate between incremental backups
  • Expected backup duration being attempted
  • Average file size
NOTE: To obtain the sizing tool, contact Customer Support.

For example, for an asset with 50TB data spanned across 50 million files being backed up using a single network port, a 36-hour backup window needs approximately three NAS proxy engines. A 3% data change rate for the asset would result in incremental backups taking approximately 6 hours leveraging the same proxy engine count.

Similarly, for an asset with 100TB data spanned across 100 million filed being backed up using a single network port, a 42-hour backup window needs approximately seven NAS proxy engines. A 3% data change rate for the asset would result in incremental backups taking approximately 6 hours leveraging the same proxy engine count.

For optimum performance, use a dedicated 10 Gigabit Ethernet (10 GbE) network interface per NAS protection engine. The ESXi network stack constrains throughput. More NAS protection engines with dedicated interfaces can achieve higher net aggregated throughput when reading from the NAS and writing to protection storage.

If the entire environment is 10 GbE, including the NAS and the protection storage, network speed constrains throughput. You may find that read throughput from the NAS appliance and write throughput to protection storage are reduced in an environment with multiple NAS protection engines. In this case, increase the number of network interfaces on the NAS and the protection storage.

Asset-level parallelism balances the number of available streams across multiple assets. PowerProtect Data Manager supports multiple simultaneous backups so long as there are sufficient available compute resources. Asset-level parallelism supports up to 256 streams per asset. Set protection engine parameters for NAS asset parallel streams provides instructions to configure parallelism.


Rate this content

Accurate
Useful
Easy to understand
Was this article helpful?
0/3000 characters
  Please provide ratings (1-5 stars).
  Please provide ratings (1-5 stars).
  Please provide ratings (1-5 stars).
  Please select whether the article was helpful or not.
  Comments cannot contain these special characters: <>()\