Tech/Engineering

Optimize Storage Costs with Druva Deduplication

September 4, 2019 Joann Starke, Sr. Manager, Solution Marketing

Ensuring continuous availability of your business’ data is a challenge. Exponential data growth, storing multiple data copies, and shrinking recovery windows were placing enormous pressure on IT. Help arrived in the form of data deduplication performed on hyper converged platforms, backup appliances, and cloud-based storage.

The growth of data hasn’t slowed and as a result data deduplication performed on hardware has made data availability complex and costly. Complexity comes from multiple server consoles and dedupe indexes that need to be managed and rationalized for a comprehensive view of your data. Then there’s the cost of storage. Hardware-based dedupe needs to store data before chunking and hashing is performed. Skimp too much on the storage cost, and the subsystem may not keep pace with the data flow resulting in a performance bottleneck.

Deduplication is one of the ways Druva has transformed data protection. Using source deduplication, chunking and hashing occurs at the source while the dedupe index is built and stored natively in the cloud. Watch this video to learn more:


Druva globally dedupes on the client (or source) by chunking files, database dumps, etc. into a slice of data that is run through SHA1 to create a hash. Comparing hashes allows Druva to determine whether the specific chunk is unique. Unique chunks are sent to Amazon Simple Storage Service (S3) and stored as an individual object. Non-unique chunks result in a hash table update.

The size of the data chunk being deduped is important because it impacts the number of records in the dedupe index. Druva uses 64K or 1Mbyte chunk sizes, which reduces the number of records stored, faster deduplication times and reduces the amount of metadata being stored. Chunks older than 90 days are automatically tiered from S3 to Amazon Glacier, which is a cost-effective way to maintain data for compliance.

How do you benefit from Druva’s global deduplication?

  • Less network congestion resulting in faster backups and restores
  • Single dedupe index ensures only a single copy of your data is maintained
  • Larger chunk sizes reduces amount of metadata and records while increasing dedupe performance

Druva customers are charged for the amount of storage they consume so these benefits translate into storage cost optimization. Saving money is always a good thing.

Go deeper and learn more about Druva’s global deduplication.