I currently use amazon ses, they are very cheap (0,1$ for 1000 messages), they hide where the mail is originally coming from and they are very reliable. However you need to achieve a low bounce rate,

(Quote)

@AThomasHowe they are indeed expensive, however I think such devices will drop in price over the next year or so. I only bought it as I had £70 worth of Amazon vouchers and only had to spend £50 or so

RamNode, SecureDragon, InceptionHosting, DigitalOcean, Vultr, DrServer, Prometeus, Serverian's brands (VPSDime etc), INIZ, Linode, Amazon, Microsoft.

(Quote)

What about serving the files from Amazon S3? You would need to use their calculator to estimate your bandwidth costs, but you could then scale your bandwidth requirements without much thought.

Amazon's AWS is compliant with dozens of standards. Amazon can host government sites, financial information, and health care records. I know of no LowEnd provider who has these certifications.

(Quote)

May the famous brands like Amazon, Windows Azure, etc will instead of all personal and little company providers? What do you think?

To be precise, this is an API driven panel that is both compatible with EC2-based API [ which includes most of Amazon's services ] as well as the OpenNebula XMLRPC2 API as described in their documenta

(Quote)

Doesn't sound like you really need a traditional dedicated that is on 24/7/365. Would a more "cloud" based solution like Amazon EC2 or similar work? Start it up, do the compile, mo

If you don't access the data often, 100GB storage on Amazon Glacier is $1USD/month.

So it has more than 150K IOPS? or 260K if you're referring to Amazon.

We disabled caching on sequential operations. So it only uses SATA for that, then it uses an SSD array for random I/O. It looks like Amazon does something similar above.

(Image)

Use spot instances on amazon ec2. Bonus is you can run multiple encoders if you have a whole set to get through fast.

First, the fact Amazon provides or does not provide self-healing, doesn't make the two pay-per-hour providers look better or worse. They just do not provide means to recover node, if hardware failure

Well, Amazon claim to have "self-healing" and still needs to reboot your VPSes?

Depends on your needs really and the devops capacity you have. Amazon, Rackspace, HP, Azure, etc are geared towards large-spend clients that have in-house expertise to orchestrate multiple VMs. Thos