Like the old saying goes, “Time flies when you’re having fun.” I am not sure whether it has all been fun, but I have been extraordinarily busy for the past few months…as my absence here might suggest.
There were ups and downs as 2016 came to a close. We have had two really good day-long (or many hour long) web-based workshops with 1105 Media and Virtualization Review covering Data Protection and Disaster Recovery in the Cloud Era — based on the Data Management Institute’s Certified Data Protection Specialist™ (CDPS™) courseware — and another covering Cognitive Data Management — based on DMI’s Certified Data Management Professional (CDMP) certification training. To our delight, each workshop had over 800 registrations and up to 300 attendees for the full expanse of the training program. We are working on two more for presentation in the coming months: one in April on Data Protection and Data Security, the second on Data Archiving. Watch this space.
We had some challenges too in these recent months. For one thing, a UPS exploded in my office, starting a small fire and spraying walls and desks with lead acid. Does anyone here know what a couple hundred TBs of storage weigh? I threw out my back lifting and hauling servers and storage into another room so we could get things cleaned up, repainted, etc. I also took the opportunity to begin redesigning my websites. Data Management Institute (datainstitute.org) and Toigo Partners International (toigopartners.com) should be up and running by end of month.
I was gifted a couple of decommissioned arrays and worked through the holidays to rebuild them, then to virtualize them with DataCore SANsymphony-V. We are still plodding along in that endeavor, owing to the need to migrate older data and to find controllers for some of these arrays. eBay has been a treasure trove.
Of course, the biggest setback was the loss a few weeks ago of one of my mentors and friend of this blog, Ziya Aral. Ziya and I have been acquainted since 1997, when he helped to form DataCore Software. As engineers go, he was one of the best — delivering technology solutions that were ahead of their time and reviled by long-standing fixtures in the storage hardware world. To his credit, not only did he presage the current fascination with software-defined and hyper-converged storage, but he also rocked the storage world with his innovative adaptive parallel I/O technology that made commodity kit run like a supercomputer. Ziya was one of those guys who wasn’t just about the benjamins; he actually had the needs of consumers in mind and never forgot his roots. I am missing him badly and wish his family, and his many friends, well as they work through their grief.
Ziya would not have wanted me to end this post on a somber note, so here are some points that are raising my spirits. First, I like the way that many companies have been deconstructing the idea of software-defined storage. Ziya and the gang had already done violence to the VMware/Microsoft ideas of what SDS storage was by including functionality in their SDS stack for RAW IO acceleration (their aforementioned adaptive parallel IO technology) and for virtualizing storage mount points. These two innovations allowed their technology to be completely hardware and hypervisor agnostic.
Acronis came to market a month or so back with an announcement of their own SDS stack that also features BlockChain support. BlockChain is one of the coolest ideas I have heard in the cloudy world in some time. IBM is working furiously on it — which is a big part of the reason why I will be going back to IBM Interconnect in Vegas at the end of March. Some of you should register for that event if you have time. It is well worth the cycles. Click the graphic to register and look for me outside the venue shooting more video interviews with IBMers.
I have also become very enamored with Strongbox Data Solutions (SDS) out of Canada. Their cognitive data management play, StrongLINK™, is way ahead of competitors and I really like what they are doing to bring capacity utilization efficiency to storage infrastructure. We need to get a lot better at managing bits if we are going to cope with the deluge of them — 10 to 60 Zettabytes worth — in a matter of three years. David Cerf and his crew are doing a yeoman’s job of creating an Uber-controller for unstructured data that works across all storage types and all file and object types. Very cool. I will be writing a lot more about it shortly.
Finally, I have recently been exposed to a new hardware technology that I see as a potential game changer for servers and storage — quite possibly the first intelligent hyper-converged cobble in the market. The product is called Iris™, it is from Symbolic IO, and I will post about it next.
Back to the grind.