Tl;dr – The “so-called supercloud” still isn’t
After taking a few well-deserved months to regroup and potentially formulate a cogent definition, the “supercloud” squad is back, declaring (yet again) “supercloud is becoming a thing”.
While that evidently was not enough time to resolve whether “supercloud” takes a definite or indefinite article, the update musters the latest case for why they hope someday to find a pony named “supercloud” amidst the mountain of media amassed around the “so-called supercloud” (their words).
Deviating from the previous strategy of writing voluminously while neglecting to offer a definition of “supercloud” specific enough to identify such a beast in the real world, this update draws inspiration from the cut-and-paste style of the classic ransom note. It is a palimpsest assembled from a smorgasbord of random and/or self-promoting vendor comments. (Note very few of them endorse or even actually use the term “supercloud”).
As we have come to expect, the latest missive doesn’t attempt to answer any of the myriad questions raised in previous discussions of the “supercloud” (such as these, these, these or these, which plead for a precise definition, examples of what does and does not qualify as “supercloud”, and how it relates to existing nomenclature and taxonomies).
Turning to the specifics of the post, we learn “supercloud” remains, at best, embryonic, as acknowledged by the title.
To marshal evidence it is somehow “becoming a thing”, the piece strings together a set of excerpted quotes much like an ancient seer might manipulate chicken entrails to foretell the future, including:
- An academic presentation from 2016 that uses the term “supercloud” to describe moving low-level compute jobs between regions or clouds. Perhaps “was becoming a thing (but didn’t)” might be more accurate? But a slow burn in the best case.
- Someone talking about BIOS and UEFI, which needless to say screams “higher level of abstraction”.
- Commentary from Dell and HPE, despite it being hard to imagine anyone more distant from what happens in a layer purported to reside above IaaS/PaaS except perhaps the vendors who lay the concrete for data center floors.
- And an IBM guy pretending that Red Hat is somehow still a thing and not a thing IBM paid $34 billion to buy
The conclusions are, to be charitable, left for the reader to discern.
But the biggest development in this installment is that the term “supercloud” itself may be in jeopardy. Previously, the hallowed term was all we had, absent any meaningful definition. We have been repeatedly promised it would “become a thing”. Now we learn its creators may not go to the barricades on behalf of “supercloud”, admitting “…it doesn’t matter to us what you call it” and “What really interested us here is not just the title… but the notion that it really doesn’t matter what it’s called.”
So, to recap, we still don’t know what “supercloud” is AND it doesn’t matter what we call it. Epistemically, what is left of a purported thing that lacks both definition and name?