How Google Compute Engine stacks up to Amazon EC2

Re:How Google Compute Engine stacks up to Amazon EC2 by oijaf888 on Friday March 15, 2013 @08:54PM (#5383882)

So your total spend on 1 reserved EC2 m1.medium instance is $579.92 a year, not including any bandwidth or EBS storage. On the other hand you could go to OVH and get a dedicated server with 8gb of ram and 2TB of disk and significantly faster processor for $468/year and that includes 5TB of bandwidth a month. (source: http://www.ovh.com/us/dedicated-servers/kimsufi.xml)

Seems like unless you are going to need the ability to quickly scale the number of instances up and down it makes more sense to put your core services on dedicated servers rather than in the cloud.

[ Parent | on HN ]

Re:How Google Compute Engine stacks up to Amazon EC2 by coldtea on Saturday March 16, 2013 @07:47AM (#5385148)

>Seems like unless you are going to need the ability to quickly scale the number of instances up and down it makes more sense to put your core services on dedicated servers rather than in the cloud.

Which is almost always the case for startups (needing to scale at will).

[ Parent | on HN ]

Re:How Google Compute Engine stacks up to Amazon EC2 by oijaf888 on Tuesday March 19, 2013 @10:16PM (#5405177)

I don't know too many startups that need to scale in minutes as opposed to hours. Even sites that have taken off have usually had growth over a period of months or years. I do think it makes sense if you are doing something massively cpu intensive that has hugely varied demand (video encoding, etc).

[ Parent | on HN ]