I started this blog over two years ago to focus on the criticality of data protection and specifically data recovery. While technology continues to evolve, the importance of these two elements remains consistent. Every company must have a recovery strategy to protect against data loss or corruption. Some people may be inclined to de-emphasize backup and recovery based on the faulty assumption that today’s virtualized hardware and software is more reliable or flexible, but this is a mistake. In the last month, we have seen two examples of why data recovery is critical, and both affected entities had large IT staffs and huge budgets. Without an effective protection strategy, massive data loss would have been unavoidable in both cases. The companies recovered the vast majority of their data but experienced an outage that was far longer and more damaging than either anticipated.
George Crump posted an article over on Network Computing discussing why storage is different for data protection. He makes a number of points regarding the benefits of using a storage appliance approach versus a software-only model, and for the most part, I agree with his analysis. However, there is an important point missing.
The software-only model relies on a generic software stack that can use any hardware or storage platform. This extreme flexibility also creates extreme headaches. The software provider or ISV cannot certify every hardware and environment combination and so the customer is responsible for installing, qualifying and testing their system. Initial setup can be difficult, but support can be even harder.
What happens if the product is not performing? The support complexities become difficult. Do you call your software ISV, your storage vendor, your SAN provider, your HBA vendor? There are a myriad of different hardware pieces at play and the challenge becomes how to diagnose and resolve any product issues. This is less of a problem in small environments with simple needs, and rapidly becomes an issue as data sizes grow.
Currently the twittersphere and blogosphere is actively discussing Fibre Channel over Ethernet (FCoE). The conversation was triggered by a post by Hu Yoshida from HDS, and I wanted to share my thoughts.
One of the most interesting responses was this one by Nigel Poulton where he explains the infrastructure required for FCoE. He goes into great detail highlighting the lossless nature of FCoE and the required hardware and cabling. The key takeaway is that FCoE is not iSCSI; it does not use generic 10 Gigabit Ethernet (10 GigE) hardware. You will need different cabling and advanced switches to meet the more stringent demands of FCoE.
As I have posted before, IBM/Diligent requires Fibre Channel drives due to the highly I/O intensive nature of their deduplication algorithm. I recently came across a situation that provides an interesting lesson and an important data point for anyone considering IBM/Diligent technology.
A customer was backing up about 25 TB nightly and was searching for a deduplication solution. Most vendors, including IBM/Diligent, initially specified systems in the 40 – 80 TB range using SATA disk drives.
Initial pricing from all vendors was around $500k. However as discussions continued and final performance and capacity metrics were defined, the IBM/Diligent configuration changed dramatically. The system went from 64TB to 400TB resulting in a price increase of over 2x and capacity increase of 6x. The added disk capacity was not due to increased storage requirements (none of the other vendors had changed their configs) but was due to performance requirements. In short, they could not deliver the required performance with 64TB of SATA disk and were forced to include more.
The key takeaway is that if considering IBM/Diligent you must be cognizant of disk configuration. The I/O intensive nature of ProtectTier means that it is highly sensitive to disk technology and so Fibre Channel drives are the standard requirement for Diligent solutions. End users should always request Fibre Channel disk systems for the best performance and SATA configurations must be scrutinized. Appliance-based solutions can help avoid this situation by providing known disk solutions and performance guarantees.
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.
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.
Scott from EMC has challenged SEPATON’s advertised performance for backup, deduplication, and restore. As industry analyst, W. Curtis Preston so succinctly put it, “do you really want to start a ‘we have better performance than you’ blog war with one of the products that has clustered dedupe?” However, I wanted to clarify the situation in this post.
Let me answer the questions specifically:
1. The performance data you refer to with the link in his post three words in is both four months old, and actually no data at all.
SEPATON customers want to know how much data they can backup and deduplicate in a given day. That is what is important in a real life usage of the product. The answer is 25 TB per day per node. If a customer has five nodes and a twenty-four hour day, that’s 125 TB of data backed up and deduplicated. This information has been true and accurate for four months and is still true today.
In this post, I highlighted SEPATON’s S2100-ES2 performance both with and without deduplication enabled. In a comment, I had also indicated that we would be adding additional performance information to our website and collateral and am happy to report that the update is complete. You can find our deduplication performance numbers on multiple different locations on SEPATON’s website including:
These documents now highlight per node deduplication performance of 25 TB per node per day.
Happy New Year!
The SEPATON S2100-ES2 was designed for speed. Our solution is based around the concept of a unified appliance which provides one GUI for managing and monitoring all embedded hardware. We also automate the disk provisioning and configuration to provide consistent scalability and performance. The result is an appliance that can easily be managed by an administrator who understands tape and wants to avoid the traditional complexities of disk.
Our performance is quite simple to understand. We use a grid architecture which means that all nodes can see all storage and can access the same deduplication repository; you can backup and restore from any node. Today we support five nodes with DeltaStor while the VTL supports 16. We will be adding support for larger node counts in the near future. Each node provides an additional 2.2 TB/hr ingest and 25 TB/day deduplication performance. The appliance deduplicates data concurrently that means that backups and dedupe occur simultaneously with no performance impact. Let’s look at the actual numbers.
Howard Marks recently posted an interesting article about NEC’s HYDRAstor over on his blog at InformationWeek. He discusses the product and how the device is targeted at backup and archiving applications. He makes some interesting points and mentions SEPATON. I wanted to respond to some of the points he raised.
…[the system starts with] a 1-accelerator node – 2-storage node system at $180,000…