Mozy is backup made easy. Create an account, download the client, select the directories you want backed up and then just leave it in the background. On a fast line I had my 2gb free account filled up in a few hours. Restoring files is easy; just request a package of files you want restored and a few minutes later you are emailed a link to download the package.
The unlimited version is $4.95 per-month which is peanuts really.
One thing to note is that this isn’t an Amazon S3 competitor. I can’t share my Mozy backed-up files with anyone else, I can’t even view a file “hosted” by Mozy. It is a backup service, nothing more and nothing less. For this it is much cheaper than Amazon S3.
One missing feature in the beta Mac OS X client I am using is external HD support. It is available in the Windows client but is “a few months away rough guess” in the Mac OS X client. Most of the files I want to backup are on an external HD so I will just have to wait. But the test I did with 2gb worked very nicely so I am confident the 150gb on my external HD will work fine too.
I just backed up the 5.5gig of my music collection to my Amazon S3 account using S3Fox Organiser. It was largely a painless process except for any song that had a # character in its name. Not sure why S3 won’t accept that but a simple rename sorted it.
I especially liked how the transfers are atomic. If for some reason a file failed half-way then it wouldn’t leave a half-uploaded file on S3. This made transfering 5.5gig painless with no worrying about what was and wasn’t uploaded properly.
Total cost? The initial 5.5gig transfer will cost me about one buck ten cents. From there on it will cost just under a dollar per month. Fantastic!
If you thought Amazon’s S3 or SQS was interesting then Amazon Elastic Compute Cloud (EC2) is going to blow you away.
Amazon EC2 presents a true virtual computing environment, allowing you to use web service interfaces to requisition machines for use, load them with your custom application environment, manage your network’s access permissions, and run your image using as many or few systems as you desire.
Each instance predictably provides the equivalent of a system with a 1.7Ghz Xeon CPU, 1.75GB of RAM, 160GB of local disk, and 250Mb/s of network bandwidth.
The images mentioned are stored on S3 (penny drops.)
This is the link between S3 and SQS I’ve been waiting for (without really knowing it.)
If I have my sums right you could run one server doing 1gigabyte an hour for 365 days for only $2,628. AFAIK that is incredibly cheap (for the bandwidth being used.)
I’ve got my beta account and am going to get into EC2.
S3Fox is an extension for Firefox that gives you an FTP like view of your Amazon S3 account. You can upload and download files, view their URLs, create “directories” (S3 doesn’t technically support directories within buckets and so this has to be faked) and set the Access Control List on items.
So the proposal is that Amazon S3 should have the equivalent of premium-rate phone lines. Since Amazon calls its fundamental unit of data segregation a “bucket”, the obvious term is “golden buckets”.
Now I am in agreement that the user should have an account with Amazon and not with Tunesafe. I believe the user should have his online storage and then allow the apps she wants to access it. Apps shouldn’t have disparate storage silos.
However I am not quite convinced about the golden bucket idea. How transparent to the user would it be? I would be on the “full disclosure” side of things but I imagine some users would wonder just why you are getting a continuing cut of what they are paying to Amazon when you aren’t providing a continuing service (you have given them the app, no more involvement from you.)
It really is a golden bucket as you could provide the app, distribute it widely and then go out of business and yet still recieve revenue from Amazon. Users might not like that idea.
S3 is fantastic but I see some serious challenges in monetizing it directly. Using it as SmugMug does is grand but with TuneSafe and similar apps there is a problem. I hope some smart business chap comes along and shows us the way.
Something I’d like to see is a Subversion repository stored on S3. From the little I know of how Subversion stores data I don’t think it would be that easy. You wouldn’t want svn sending the entire single repo file up and down to S3 everytime. You’d want the individual items and their deltas being sent.
Amazon have released their Simple Queue Service, SQS, to the world. Technically it is exactly what the name suggests; a simple queuing service. Like S3 it does one thing and it does it well. Like S3 pricing is on usage, 10 cents for every thousand messages and 20 cents for every gig of data transfered.
Effectively though it is a lot more than that. Tie SQS up with S3 and some simple Rails code and you have a massively scalable, reliable application.
They also provide a Windows Communication Framework “add-in” which should make it really interesting for Windows developers.
I am really liking what Amazon are doing here, taking the core technologies found on servers that power applications, making them globally available and providing the kind of power smaller developments didn’t have access to.
Mozy is backup made easy. Create an account, download the client, select the directories you want backed up and then just leave it in the background. On a fast line I had my 2gb free account filled up in a few hours. Restoring files is easy; just request a package of files you want restored and a few minutes later you are emailed a link to download the package.
