Attention IBM i Users: You Deserve Better Backup!

By Caitlin Gordon, Senior Product Marketing Manager, EMC Backup Recovery Systems

Why should open systems environments have all the fun? Your IBM i environment deserves better backup too and EMC Data Domain deduplication storage systems can help you by eliminating:

  • Tape handling
  • Offsite tape shipping and storage
  • Failed recovery due to bad tape media
  • Tape refresh costs

So what’s the catch? Like many others before you, you’re probably thinking that this solution will be much more expensive than tape.  It’s certainly the first question that most customers ask when discussing the idea of replacing tape with a Data Domain system.  The good news is that Data Domain systems enable a Total Cost of Ownership equal to or lower than tape in most environments.  So what’s the secret?

The key to enabling this cost effective next generation backup solution is deduplication technology.  Deduplication enables you to significantly reduce the amount of backup storage required to protect a given data set.  For example, let’s say you have a 1 TB IBM Power System that you backup with a traditional weekly fulls and daily incrementals. Without deduplication, in one full week, you’ll have backed up (1 TB + 50 GB x 7 + 1 TB) 2.2 TB of data, but this has a lot of redundant data in it, so without deduplication you’re wasting precious storage and floor space.

Deduplication changes the game by enabling you to store only unique data.  As a backup comes into the Data Domain system, it is split into small sub-file segments that are analyzed for uniqueness.  If the system has already stored that segment from a prior backup, it will not store the data, but will insert a pointer instead.

The chart below shows how this deduplication technology results in significant backup storage savings for the environment we discussed above.  Specifically, when the first full backup is done to a Data Domain system, you’ll find a bit of redundant data within that backup and achieve an average of 2 to 4x data reduction.  For a 1 TB first full backup, a 4x data reductions means that only 250 GB is required to protect that 1 TB backup.  Similarly, each 50 GB incremental backup will contain data that was already stored in previous backups, resulting in 10x reduction and therefore only requiring 5 GB to protect.  However, the biggest savings comes in the second full backup when the vast majority of the data has already been backed up, so only 18 GB are required to protect a 1 TB backup.  So you can see over time you backed up the same 2.2 TB, but it only required 288 GB to protect – an overall deduplication ratio of 7.6x.However, this is just the beginning – the most impressive backup storage reductions happen as you retain full backups over time.  The chart below illustrates the cumulative deduplication ratio that can be achieved over a 3-month period.  The first two rows represent the first and last row of the chart above – showing a ~8x reduction after 1 full week.  However, as time goes on, there is even more redundant data being backed up and the result after 3 months is a cumulative 21x deduplication ratio where only 706 GB is required to protect 15.4 TB of backup data.

This type of storage savings is only the beginning of the story.  The other key to deduplication technology is how it affects your DR strategy.  With deduplication, it is now possible to replicate data to an offsite facility without requiring expensive network upgrades.  This is accomplished by only sending unique data to the remote site – enabling up to 99% reduction in bandwidth required for fast ‘time-to-DR’ readiness.

In addition, these systems easily integrate with your existing environment by enabling backup via BRMS to a Data Domain system instead of tape (stayed tuned for more on this in April).  When you combine this ease of integration, storage and network efficiency with the ability to eliminate the risk and costs of tape, you can why deduplication has transformed backup in open systems environments.  Now it’s your turn – to learn more about implementing a Data Domain system in an IBM i environment, check out this solution overview.

Caitlin Gordon

Caitlin Gordon

Data Domain Product Marketing, Data Protection and Availability Division
I have spent the past nine years focused on all things data protection, with a focus on backup and archive. Lucky for me, Data Domain Systems give me lots of good topics to discuss here. When I’m not blogging, I’m equal parts gadget geek and sports freak – always ready to chat about the latest IT rumor or celebrate/lament the latest Boston sports heartbreak/victory. You can also find me talking backup on Twitter and YouTube.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>