Part of the flip side of computing becoming a true utility service is that the infrastructure providers will have massive amounts of hardware, in order to drive the efficiency of the marketplace, a clearinghouse where vendors can unload excess capacity at discount rates is necessary
My friend and sometimes colleague Krishnan also reviewed the service and, while generally positive, raised a few questions around the lack of transparency with SpotCloud and in particular the fact that customers have no idea from whom they are purchasing service, and no SLAs protect the service they’re buying. All would have ended there but both Krish’s and my own comments were picked up by the MIT technology review in this article. The article selectively quoted from Krish and gave the perception that he was generally down on Sportloud. Zoli Erdos, Editor of CloudAve (and another sometimes colleague of mine, the cloud community being surprisingly small) came to his defense criticizing the selective nature of MIT’s quotes.
Egos got bruised and misunderstandings eventuated which have now, thankfully, been soothed. But a core issue stands, outside of how Krish’s comments were used – that is the issue of whether or not Qqality and transparency of service is really an issue in a clearing house situation. I told Krish that I believe specifically in these sort of low-cost, capacity clearing situations, customers expect an approach which is more akin to an “as is, where is” transaction than a boutique level of service and quality. Put it this way, if I go to a discount mart, looking for a $5 pair of jeans, I’m going to be pretty relaxed if, when I get them home, they’ve got fabric flaws or whatever. If I go into a boutique however and buy a name brand pair of jeans – I’m going to expect a level of quality significantly different.
Krish misunderstood my comments and assumed I was suggesting that enterprises have no requirement for transparency, of course they do, risk aversion is in their blood. Fact is no enterprise is going to use SpotCloud for their mission critical workloads, what they may do however is use SpotCloud for some unimportant tasks, off to the side, where the odd outages are irrelevant. However Christian Reilly from Bechtel responded to my Focus.com question with an interesting answer, saying;
One of the great unspoken worries around public cloud is the performance (and cost of transfer in/out) of the network that carries the traffic to and from the provider. There are some concerns that even the “leading” IaaS providers don’t have particularly great networks, in terms of peering relationships, and so the question of whether I would accept lower stability or quality would depend on the workload. If I used SpotCloud to run a dev project, I’m inclined to care less about the network performance than I am the stability of the kit under the application, but in most production cases, I don’t think anyone in their right mind would want to forego network performance for cheaper compute and storage.