Today EMC introduced the industry's fastest deduplication storage systems. And they are fast. Big and fast. Impressively so, and even more impressively if you have been involved with backup systems for a while—in a "look how far we have come" sort of way.
But to focus on just the speed and scale would be to miss a lot of great functionality that our engineering teams have worked very hard to include in the latest software and hardware release of the Data Domain systems. In fact, this may be one of the biggest releases ever for new functionality and new capability with the release of Data Domain Operation System (DDOS) 5.0.
Before I go there however, let me spend some time wowing you with just how big, and how fast, Data Domain systems have become. How about 175 times faster than the first Data Domain system introduced in 2004. How about 450 times more capacity than what was available 7 years ago? A little while ago I asked the question: "What curve are you on?" The answer to that question is all about the relevance of increasing the speed and capacity of backup systems at a rate that is at least equal to the rate of data growth for most organizations. Simply speaking, if backup systems are not capable of keeping up with data growth, the architectural model of those backups systems has a notable weakness. At very least, the number of those systems required to keep up with your data growth will increase over time; adding cost and complexity to the environment. Exactly the wrong direction to be moving in.
So if it seems like we spend an inordinate amount of time on this notion, there is a very good reason why: it is only through speed and capacity increases like EMC has been able to achieve with its Data Domain systems that we can move in the opposite direction. The direction of growing performance faster than data. By growing performance and capacity faster than organic data growth, we are actually reducing the number of systems that a given environment will require year over year for backup purposes.
So just how big and how fast are the new systems? The newly introduced DD890 now scales to 14.7 TB/hr backup ingest, and up to 285 TB of useable capacity. The Data Domain Global Deduplication Arrary (GDA) scales to 26.3 TB/hr and up to 570 TB of useable capacity. For the average customer, that would equate to the ability to backup more than 200 TB in an eight hour backup window, and retain more than 10 PB of data.
What is more is that the GDA has made a major step forward in terms of usability and broad market appeal, by adding support for the Data Domain VTL software option in DDOS. Previously, if you wanted to deploy a GDA, you had one choice on what protocol to use: DD Boost. Now that has changed, and the GDA is useable with either DD Boost or FC (VTL) interfaces. As a result, it is useful for TSM environments, and almost every other major backup application. In my experience, this is doubly important because it is those big IT environments that are likely to be interested in the GDA that are also most likely to be heavily invested in FC infrastructure for backup.
Two other major connectivity improvements have been made with this latest release of DDOS too, and are therefore generally available for all Data Domain systems: IBM i connectivity and NDMP connectivity. With DDOS 5.0 we are now supporting the direct attachment of IBM Power Systems systems running IBM i to Data Domain systems. Those systems can be shared between IBM i and open systems too, so no longer does IBM i need its own dedicated (and costly) backup infrastructure. And it can leverage the network efficient replication capabilities of the Data Domain system to provide for fast, and cost effective, disaster recovery.
Also new on the connectivity front is the addition of NDMP over Ethernet functionality. Say goodbye to redirected NDMP backup streams: the Data Domain system can now be a direct target, over IP, of the data path backup data of an NDMP backup. Again, the functionality works with all major backup providers and NAS systems from both major providers.
Then there are a host of lesser improvements in DDOS v5.0 that contribute to this being a very significant release indeed. Some of these are worthy of further discussion—and I will do just that if questions or comments arise on specific capabilities—but briefly the new functionality includes:
- Enhanced replication management
- Enhanced reporting capabilities
- Role based administration/access control
- Improved remote management
- An improved configuration wizard
- All new functionality in DDOS is available via the CLI
- Enhanced SNMP functionality DD Boost with encryption for encrypted optimized replication with BE, NBU, and NW
- DD Boost load balancing and link failover enhancements
- Encrypted replication
- LACP support for increased performance
- Myriad and sundry smaller hardware, connectivity, and OS enhancements
All in all, this is a very significant release, and a very significant step forward not just for performance and capacity, but for the DDOS itself.