Day 3. With the finishing line in sight, the day started with a Storagetek partner panel, followed by a customer panel and a presentation of Dan Berg, VP and CTO in EMEA.
I passed on the first slot of the breakout sessions today, to spend some time with my Sun storage peers. We had a great and sometimes heated discussion around the topic of ISVs in the context of Sun's storage business. We also compared notes on our lifestyles, especially interesting since we had participation from the UK, Germany and the US. Good fun, we laughed a lot. Thanks, Tim and Stefan.
My next presentation was on the capabilities of Sun's high-end disk array offering, the Sun Storagetek ST9900 family. This is a product Sun brings to market in partnership with Hitachi. The latest incarnation of this product is the Sun Storagetek ST9990V. The “V” represents the virtualization capabilities of this new model. In laymen terms, this allows you to divvy up a single ST9990V to make is look like distinct, independent disk arrays. One used case for this was a bank that wanted to separate mission-critical data from the rest of the business.
But it was two numbers that caught my eye. The ST9990V is capable of 3.5M IOPS, and with 300 GB drives can reach a maximum capacity of 332 TB. Wow, that's a lot of storage.
The last session of the day was presented by Chris Wood, CTO of Sun's storage practise. Chris managed to put my head straight around the terms of data availability and data protection.
Chris defined “Data Availability” as : “data is accessible to the application whenever needed and at the required performance”. That obviously means that just because data is archived on tape does not mean it's available. It might rest safe and secure in a vault, miles away from any tape drive.
On the data protection, he warned the audience about jumping to the answer to quickly. He urged us not to jump to the product pitch too quickly, and instead ask our customers what they want their data to be protected from. Is is data loss, corruption, hackers, operator errors, bit rot ? Depending on the needs, your answer might be very different. I remember Chris using an example where a customer used tape backup to prevent data loss, yet a recent data corruption meant that the customer was backing up bad data again and again.
On the topic of data loss, Chris had a great sound bite, I've heard before, but I still liked it. It goes as follows. There are only two types of disks : Those that have failed, and those that are about to.
Chris had some excellent customer examples. One of which, he demonstrated how a customer used SAMFS to consolidate their backup strategy onto two SL8500, and reclaim 21% of floor space in their data centre. Cool stuff.
That was the last session. Storage Academy 2007 is over. I had a great time, and can't wait for next years event.