About Restore

Recently an end user commented about how the replication performance on his DL3D 1500 was less than expected. As he retained more data online, his replication speed decreased substantially and EMC support responded that this is normal behavior. This is a major challenge since slow replication times increase replication windows and can make DR goals unachievable.

The key takeaway from the comment is that testing is vital. When considering any deduplication solution, you must thoroughly review it with limited and extended retention. In this case, the degradation appeared when data was retained and would not have been found if the solution was tested with limited retention. The key elements you should test include:

1. Backup performance
   1. On the first backup
   2. With retention
2. Restore performance
   1. On the first backup
   2. With retention
3. Replication performance
   1. On the first backup
   2. With retention

Data Domain recently announced that their new OS release dramatically improved appliance performance. On the surface, the announcement seems compelling, but upon further review, it creates a number of questions.

Performance Improvement
Deduplication software such as Data Domain’s is complex and can contain hundreds of thousands of interrelated lines of code. As products mature, companies will fine tune and improve their code for greater efficiency and performance. You would expect to see performance improvements from these changes of about 20-30%. Clearly, if an application is highly inefficiently coded, you will see greater performance gains. However, larger improvements like those quoted in the release are usually only achieved with major product architecture updates and coincide with a major new software release.

In this case, I am not suggesting that Data Domain’s software is bad, but rather that the stated performance improvement is suspect. They positioned this as a dot code release and so it is not a major product re-architecture. Additionally, if it was a major architecture update, they would have highlighted it in the release.

To summarize, the stated performance gains in the release are too large to attribute to a simple code tweak and I believe that the gains are only attainable in very specific circumstances. Data Domain appears to have optimized their appliances for Symantec’s OST and is trumpeting their performance gains. However, OST represents only a small fraction of Data Domain’s customer base and it seems that customers using non-Symantec backup apps will see uncertain performance improvements. Read on to learn more.
Read more…

Scott from EMC posted about the EMC DL3D 4000 today. He was responding to some questions by W. Curtis Preston regarding the product and GA. I am not going to go into detail about the post, but wanted to clarify one point. He says:

Restores from this [undeduplicated data] pool can be accomplished at up to 1,600 MB/s. Far faster than pretty much any other solution available today, from anybody. At 6 TB an hour, that is certainly much faster than any deduplication solution.
(Text in brackets added by me for clarification)

As recently discussed in this post, SEPATON restores data at up to 3,000 MB/sec (11.0 TB/hr) both with deduplication enabled and disabled. Scott insinuates that only EMC is capable of the performance he mentions and I wanted to clarify for the record that SEPATON is almost twice as fast as the fastest EMC system.