On the Threat from Efficient Cloud Computing

March 26th, 2009 by · Leave a Comment

On Seeking Alpha yesterday, Alex Salkever posted an interesting article discussing how efficient cloud computing may change the sector.  It almost certainly will, and I'd like to add my two cents on the subject.  Specifically pertaining to datacenters, he said:

"Data center operators like Equinix (EQIX) could also feel the pinch. They all expected massive demand for custom designed homes that featured servers and other critically important hardware used to run networks and businesses. But if there is less need for servers, there is less need for data centers."

While there is certainly a kernel of truth in that, there are several misconceptions that I would like to address.

The first is that it implies there is some sort of datacenter glut going on, that the sector has overbuilt.  The build-out in datacenter space over the last few years was not based on hopes, it was based on sales orders.  Most of that space is already spoken for, and there is something of a land rush going on for the remaining space - even that which is incomplete.  If efficient cloud computing becomes a real and pervasive phenomenon, it will simply help us all avoid the pricing train wreck that might otherwise occur due to the negative effect of the credit crisis on new datacenter construction.

Secondly, we should recognize that this is not a simple case of cause and effect - cloud computing is not an external force, it is economically linked to datacenter and colocation pricing and products.  To a large extent, it is the very scarcity of datacenter resources that is one of the major forces driving the cloud computing effort.  The costs to house, power, and operate massive service farms hav been rising, and it is those costs which give cloud computing the economic basis for its development.   If rapid deployment causes a drop in datacenter pricing, it also reduces its own rationale for deployment.

Finally, it is easy to suggest that less space and power required is a threat to providers like Equinix (NASDAQ:EQIX, news, filings) or sdxc, because those are two of the more visible, raw, physical products sold.  But this misses the point that it is not space nor power that a datacenter can provide better than anyone else, what Equinix and others like them provide is actually centralization of resources and access to bandwidth.  Cloud computing HELPS them do this better.  It helps them scale their facilities so they become more valuable to more customers.  They may sell less space per transaction, but that space will also be more valuable and require less capital to create relative to its value.

In fact, the biggest effect may be to breathe new life into older datacenter facilities, many of which are either full physically or constrained by power.  Another 'winner' would be the datacenter-centric bandwidth providers such as Cogent Communications (NASDAQ:CCOI, news, filings) and Level 3 Communications (NYSE:LVLT, news, filings).  After all, an efficient cloud needs less servers to do the same job but I don't see anyone suggesting that it will need less bandwidth.  Higher server utilization in an otherwise full datacenter means bandwidth must necessarily rise.  Thus, rather than build new connections into new datacenters in order to grow, such bandwidth providers might see greater growth from existing buildings at lower expense.

I'm not saying there won't be effects, there is the potential for a technological singularity here.   It's just that it is far more complex than simply one new technology affecting demand for an old product that we can then say is at risk.

If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!

Categories: Datacenter

Discuss this Post


Leave a Comment

You may Log In to post a comment, or fill in the form to post anonymously.





  • Ramblings’ Jobs

    Post a Job - Just $99/30days
  • Event Calendar