After a long flight, I decided to attend the pre-IBM Interconnect event, the Open Technology Summit (#IBMOTS) at Mandalay Bay Resort in Las Vegas. It was a good choice.
I admit to being somewhat remiss in keeping up on all the open initiatives out there — there are so many and little to guide me to the ones that are most important or significant. I understand that a lot of work is being done on platforms, on API integration, on cloud-native apps and operating systems, and so forth. But what always seems to be missing is any discussion of data, which is the whole reason for computing in the first place.
In the old days (when I was a data center newbie), we started development by looking at the process we were automating, considering all of the inputs and all of the outputs (both of which were data) and trying to document the kinds of transactions that would occur, frequency of data accesses and updates, useful life of the data, that sort of thing. Then — and only then — we would start designing the infrastructure to support the data and its performance requirements, scaling requirements, etc. I learned to do this, as did most folks I knew, from Big Blue: systems development lifecycle methodology. Oh, and we called it “data processing” — not “information technology” — because data processing was the sine qua non of corporate computing.
Okay, so those were the old days. Now we are on to cloud computing, with its many nuances and integration challenges. I get that. But the reason for doing it remains: it’s all about the data. Disappointingly, there was virtually no reference to this in the Open Technology Summit, just a lot of lip service to openness and communities and collaborative development by vendors who appear to be committed to openness only on the surface. (An engineer told me quietly that OpenStack was of interest to IBM only insofar as the OpenStack model could be ported to mainframe hosting: “You can create a cloud infrastructure out of commodity components, to be sure. But then you just have a commodity cloud, not a mainframe cloud with all of the dependability, resiliency, scalability or performance that mainframes can deliver.”)
So there’s that. Open doesn’t really mean open. Collaboration doesn’t mean marriage. I get it.
Still the lack of discussion about data grated on me. Especially since every large cloud vendor I have been talking with is ringing his or her hands at the prospect of the 10-60 ZB data deluge on the horizon. Heck, clouds need to do things cheaper, with less risk, and better in terms of performance than what companies can do in their own data centers if they are really going to take off. I wanted to hear somebody — anybody — tell me how much better the management of data was going to be in a cloud. I didn’t hear it.
Until Monday Morning. Bless his heart, IBM’s opening keynote at Interconnect was all about data first. Once Jeff Moody, GM for Twitter, vacated the stage, Arvind Krishna took the stage and gave IBM’s viewpoint. The first word on his opening slide was “data.” Integrating data, cleansing data so it could be analyzed, storing data were the key challenges of cloud, he said.
Amen, brother. Great presentations followed, each exposing another dimension of the need for data management — to protect data, to secure it, to preserve it, etc. Heavy emphasis was placed on Watson and cognitive computing to help automate the data management process and there was even a mention of a new technology offering, a Cloud Object Storage something or other, that Big Blue was pushing into the market. Little discussion of the internals, only that it was the next big thing in storing mass quantities in the cloud — or in several clouds — or on premise and in the cloud — with a common management model. Sounds good. I want details — especially how it will leverage tape, which it must if it is to handle the Zettabyte Apocalypse.
Anyway, good show thus far. Looking forward to more sessions and to interviewing the movers and shakers at IBM.
BTW, I had my registration, travel and lodging comped by IBM in exchange for live tweeting and doing some blogging around the event. However, all opinions here are my own.