I am a cloud sceptic. I will likely be ostracized by some nephophilic colleagues for saying this, but I don’t believe that I’m ever going to store all my data in the cloud and access it wherever I go. There are many reasons for this, but let’s focus on one – data transport.
Google can build clusters with hundreds of thousands of cores. But they are connected to me by a feeble DSL link. That’s about 30 Gb down-link a day flat out if there were no contention (as if). By 2012 if I’m really lucky I might have fibre to my home which in real life might deliver 0.5 Tb on a good day.
By 2012 we’ll likely have terabyte flash drives. Small ones weight almost nothing, so I could stuff a few in an envelope. So we have a packet size of several Tb and with overnight delivery that gives the postal service a data transport capacity of a few Tb a day at letter rates. That’s several times better throughput than fibre. There are several technologies that might deliver a multi-terabyte DVD by 2012 as well. And in the same time-frame hard disks will store multiple Tb, although as they’ll be a bit heavier the postage charges might be higher. But in terms of data transport capacity those cloud infrastructures won’t even be able to compete with a 300 year old postal service. Clouds won’t be fast enough to back up or deliver the sort of data volumes that persistent storage will hold.
Now this argument is maybe a bit contrived, because not everyone needs that much data and a terabyte flash will likely cost a few hundred dollars even in 2012. However, there’s a serious point here. You can’t be too rich, too thin or have too much storage. A decade ago I couldn’t imagine why I’d need a terabyte server at home, now I need to upgrade it for more capacity and better streaming performance. In a decade or two I might own a petabyte and want real time access to multiple terabytes each day. I don’t know why, but I will. And whatever happens, my data demands will grow faster than the national broadband infrastructure. So neither my bytes nor my head are going to be in the clouds. I’m hanging on to my own data.
3 responses so far ↓
1 4sysops - The ever ongoing race between cloud and on-premise computing // Jan 8, 2009 at 9:27 pm
[...] Analyst Nick Jones posted an interesting article in his blog with the title The cloud computing fantasy. Contrary to many analysts, he is a cloud skeptic. The argument he puts forward is my favorite one [...]
2 Brian Nelson // Jan 13, 2009 at 12:45 pm
Interesting post. I’ve been considering the same thing. I installed Drop.io several months ago and then promptly copied up a ton of files and yet, everything I access on a regular basis still sits on my hard drive. In fact, I’m not even sure what is and isn’t up on drop.io anymore.
Turns out that I like having MY files on MY computer, not somewhere else. So, sync services have become my use of the “cloud.”
3 Personal supercomputer wants to meet genetic algorithm for simulating relationship // May 24, 2009 at 3:40 am
[...] core data centres and we’ll never see it at home. But as I’ve mentioned in other blogs I’m a cloud sceptic and I don’t think the world will evolve that way, if only because of the imbalance between [...]
Leave a Comment