Jason Hoffman of Joyent has written an interesting article, Why EC2 isn’t yet a platform for “normal” web applications.

1. No IP address persistence (they all function as DHCP clients and are assigned an IP). One has to use dynamic DNS services for a given domain.
2. No block storage persistence. When the instance is gone, the data is gone. Yes I know you can send this back regularly to S3, but isn’t that actually a ‘hack’?
3. No opportunity for hardware-based load balancing (which happens to be the key to scaling a process based framework like Rails and mentioned above).
4. No vertical scaling (you get a 1.7Ghz CPU and 1 GB of RAM, that’s it). So like the block storage problem, this hits databases, we run about 32GB of ours in memory.
5. Can’t run your own kernel or make kernel modifications so there’s no ability for kernel and OS optimizations, and no guarantee that they’ve been done.
6. Images have to be uploaded and then moved around their network to find a launching point. This can take several minutes, if not more. Move 100 GBs around a busy gigabit network sometime and see.

Just received an email from Amazon about the pricing change of Simple Storage Service. Instead of charging a flat rate of $0.20/GB uploading and downloading, “traffic cost” will be split into two from 1^st June onwards — a per request cost and bandwidth cost. It makes sense because not every gigabyte fetched costs the same for Amazon. 1,000 requests for 1Mb object will definitely be much more expensive than one single request for a 1Gb object.

Here is the relevant section in the email:

What do you do when you have regular traffic spike? Say, for once a month, traffic increases 3 fold for 12 hours after your company sent out the monthly news letter? Your current web server barely copes with regular load. Do you go out to buy 2 more dedicated servers just for that 12 hours in a month? That wouldn’t be too economical paying 2 extra servers sitting there idling most of the time, wouldn’t it?

Judd Vinet of ArchLinux (one of my favourite, btw) has recently written an article to solve this very issue, Web clustering with Amazon EC2, where extra servers are hired from Amazon EC2 on part-time basis to serve surge in traffic. A semi-automated system has been built to make the task of “summoning new servers” much easier, and has been discussed in the article.

Amazon Web Services is expensive, if you compare them with the overselling dedicated server market.

Jeremy Zawodny’s blog post from early last month has prompted me to look at offsite backup solutions again. Currently I am backing up all my websites, from various servers and accounts, to my home server using rsnapshot, running at 4am every morning. So far so good, and I loves the flexibility of rsnapshot. I guess if one of my server dies, it would be trivial re-populating another server, moving the DNS records, and start serving again.

Moreover, the cost of running a home server is in fact less than what Jeremy has calculated. My home server (a 1Ghz Duron + 3 smaller disks) uses less juice, but more importantly, it needs to be running anyway regardless whether I am using it to perform backups or not, as it also provides a few other services. Like, acting as a file server for my home network.

I am an Australian, and Australia is a tiny place. Well, the land is big, but majority of 20+ million people chose to squeeze into a few coastal cities. Web hosting is expensive here, and everyone went “wow” when Aussie HQ introduced their million-dollar real time provisioning dedicated server this month (see all the buzz on this Whirlpool thread). I mean, you have enough customers and servers in Australia to provision dedicated servers in real time? Instead of hours or even days of waiting? No wonder they claimed to be Australia’s first.

But it is a dwarf in comparison to Amazon’s Elastic Compute Cloud (EC2), which is currently at limited beta.