Data storage trends 2011: Predictions of hot data storage technologies

Among the data storage trends for 2011 are primary storage deduplication, automated tiering and storage resource analysis software for better business analytics.

Those are some of the predictions offered by industry analysts Tony Asaro, Valdis Filks and Greg Schulz when we asked them to pick the three hottest data storage technologies for the coming year.

To get their full answers, read the podcast transcript below or listen to the MP3.

Tony Asaro, founder and senior analyst at INI Group, Wareham, Mass.The first one is file discovery analysis and management software, and I say that because the amount of files we're seeing out there today is just crazy. More and more capacity is growing. There are predictions that file capacity is going to outstrip all other data types in the data center in 2014 by a factor of 10-to-1. So, it's causing a lot of problems and a lot of issues out there. Instead of just throwing more and more storage at it, we have to have better tools that allow us to understand what we have stored from a file perspective -- what's dormant data, what's duplicate data, and how we can actually move data to lower tiers of storage for cost effectiveness and storage optimization. It has to happen.

The other one that I think is really important, and we're just beginning to see this come out now, is data deduplication and compression for primary storage. Again, the problem is that we're just consuming so much capacity. It's taking up so much floor space, that even if you can get a factor of 4-to-1 [reduction] on primary, you're going to be saving significantly. Arguably, even 2-to-1 is pretty powerful. There are a lot of challenges with doing it. You need smaller page sizes in your storage systems. You need a lot of memory. You need a lot of CPU. But Moore's Law is enabling that now so that we actually can start implementing these technologies in primary.

The third one is automated tiering, and this is at a sub-LUN level. This just means I can take stuff in a volume that is inactive and move it to lower tiers. This becomes very powerful. I've seen customers save millions and millions of dollars by doing this. The concept is this: You've got a volume; 90% is inactive; 10% is active. Keep the 90% on lower tiers -- it will save you a ton of money both on the disk drive level and at the RAID level you use -- and take the other 10% and keep it at your fastest tiers. Best price/performance. That's what everybody is looking for in the data center, and I think it's probably the most important technology in storage in 2011 and beyond because it will change the cost landscape significantly in [storage area network] SAN storage.

Valdis Filks, a research director for storage technologies and strategies at Gartner Inc., Stamford, Conn.

The first technology I believe will be "hot" is called information dispersal algorithms. This is incorporated into products produced by companies like Caringo, Cleversafe, Netgear, QNAP or D-Link, which use the BitTorrent protocol. This is true for the Internet or cloud storage, where you can have many, many small nodes in remote offices all connected by the network. When you store a piece of data, it's dispersed or spread across the network. [It's] very useful for storing objects or unstructured data. If one node fails, it doesn't matter. You can use all the other nodes to get the data back. It's a new technology, very suitable for storing unstructured data in what we want to call the cloud.

The second one is solid-state disks [SSDs]. This is, again, not a new technology. But what happened in 2010 is that most vendors incorporated solid-state disks in what's called automated tiering solutions in storage arrays, where the data is automatically moved between the disk tiers and the SSD tier transparently to the application without any operator or storage management admin. This increases the overall performance of storage arrays, and it improves the cost benefit of using SSDs, and as performance changes during the day, the dynamic performance of the system changes. [It's] very useful in situations where you have virtual desktop infrastructures.

And the last, but not least, hot technology for 2011, I believe, will be primary deduplication, or what we call data reduction technologies, which is where we can deduplicate or compress data on primary storage arrays and filers. This enables companies and organizations to store more data in the same storage array for less money. It's already available in some solutions and storage arrays and filers today. But I believe it will expand in 2011.

Greg Schulz, founder and senior analyst at StorageIO Group, Stillwater, Minn.

For 2011, the three hottest, most talked about technologies involving storage are going to be storage or systems resource analysis (SRA), business analytics, as well as data footprint reduction.

[SRA tools] give you the insight … the metrics and measurements that allow you to make intelligent, informed, smart decisions about how you're deploying your IT, your storage resources, as well as for supporting initiatives such as cloud, virtualization and other optimization-type techniques.

Business analytics: It's all about data warehouses. It's about leveraging information, transforming data into information for making decisions. But it's also about pulling in data that's large data, different types of data, extremely unstructured data to figure out trends and to spot patterns, whether it's for business decision support or infrastructure resource management decisions.

The other one is data footprint reduction; in other words, looking at how to do more with what you have, how to stretch your storage budget with things like archiving, backup modernization, compression, real-time as well as traditional compression, dedupe, thin provisioning and space-saving snapshots. [The theme is] doing more with what you have, taking your existing IT storage budget and supporting growth.