With SNW 2010 only three weeks away, my briefing calendar is already filling up, but I am trying to determine what some of the key themes are going to be this year. While we can't say the economy is rocking and rolling again, it does seem that IT Professionals are on the hunt to solve some specific challenges in their data centers. Here are my predictions of what is going to come out of SNW 2010.
First, of course, there is going to be a lot of discussion about automated tiering. One of the tools that vendors need to either develop or start promoting is something that will show customers whether or not they will benefit by putting in a tier of solid state disks (SSD), or if they can save money without sacrificing performance by moving data to a SATA tier. Now that the major manufacturers have either announced that they are actually delivering automated tiering, serious conversations can happen on the differences between the products, and what is going to deliver the best return on investment. I also think there is going to be some call from vendors that moving data to a traditional SSD tier or archive tier, manually, is more cost-effective and may deliver a higher return on the investment.
I think automated tiering becomes a no brainer for customers purchasing the solution as an upgrade. It may also become a key decision point for customers looking to replace their existing primary storage platform. I don't know if customers will move to a new storage platform prematurely to get automated tiering. This leaves the door wide open for companies that can add tiering capabilities to existing storage.
Another key theme is going to be infrastructure consolidation. Most will call this theme FCoE. Although an important component, I think it is more than FCoE. There is going to be some discussion around the role of large backbone or director class switches as well as continued discussion about I/O virtualization. Also don't leave straight 10GB Ethernet out of the conversation. There are a few makers of this high-end technology that think intelligent use of 10GBE or 20GBE is the way to go. As we begin to increase disk and network performance, there is a need for a set of tools that can provide real time measurements of how much of these high-speed devices your physical and virtual machines are using. With that insight, you should be able to better determine how much, if any, of these higher performance options you should be taking advantage of and, if you can, where exactly you should apply them. We will be meeting with a few vendors that claim to provide these capabilities.
That's just the beginning. We still have dedupe wars, SAS vs. Fibre, and of course, the great cloud storage debate to work through. We'll make some predictions about those in our next blog. Finally, we'll be there, as usual, providing briefing updates as they happen, so if you can't make it, you can tune in to get those updates throughout the day. If you're a supplier and want to get on our calendar let me know.