So given that history, it is interesting to hear of a company coming out of stealth today that is promising to deliver the first "real-time hybrid cloud software that streams production workloads to and from the clouds in minutes." While the pedant in me would grumble about the somewhat oxymoronic use of "real time" and "in minutes," I'll not quibble but will rather look at who this company is and what they do.
Velostrata is an Israeli-based startup that has raised $14 million in Series A funding led by Norwest Venture Partners and Greylock IL Partners (83North). Issy Ben-Shaul, CEO, and Ady Degany, chief product officer, founded Velostrata in 2014 with the vision of enabling real-time streaming of any workload to and from the cloud without the risk, cost, time, or complexity associated with migrating storage. According to the marketing blurb:
The idea behind Velostrata is that organizations can leave their storage on-premises in order to comply with regulatory or other requirements. Meanwhile, the actual application itself can be sited on the public cloud. Using Velostrata, organizations can stream production workloads to the cloud, maintain the location of their data, and still get good performance.
MyPOV
It seems to me that Velostrata is a few years too late. While I can accept their contention that there is some demand from enterprises to follow this dual approach towards individual applications and have compute in one location and data in another, it seems to me the world has moved on from this somewhat simplistic view of what hybrid means.
The widespread interest in containerization, and the parallel move towards using microservices within enterprise applications, has meant that this dual approach towards hybrid IT has largely been overtaken by a far more complex and wildly heterogeneous approach towards the problem space. As such, the sort of initiatives and products that will gain traction going forward are those that offer a layer of orchestration and management over complex workloads.
To put it simply, Velostrata's approach might take a company from the dark ages of enterprise IT and pull them a few years forwards, but it doesn't really allow your average enterprise to look or work like a Google or a Twitter. And yet this is what successful organizations will need to do - their IT will need to deliver the sort of flexibility and moldability that we see from the biggest web-scale operations.
Add to that the fact that we have approaches such as Microsoft Azure's stretch table database, which allows a database to stretch between on-premises and cloud locations, and we can see that this simple approach towards hybrid is rapidly becoming insufficient.
Velostrata might be delivering something pretty cool that we might have thought was impossible a decade or so ago, but the world has moved on. While this may well be useful to some organizations, it doesn't really move the needle in the way it needs to.