Does This Cloud Compute ?

Well – I see Microsoft this morning re-branded their cloud compute effort as ‘Azure’ which was previously described as being “Windows in a cloud” – hopefully one won’t accidentally open and allow someone to fall through, as the landing could be a bit rough !

Looking at the what is being presented, I would have said it looks like the typical combination of aspiration (vapour ware) and concept software (beta version) that is often presented by software companies. Microsoft has a history of of presenting ‘me too’ products to their corporate accounts and to the press, which often as much are really just an attempt to inject enough of what Gartner calls ‘FUD’ (fear, uncertainty and doubt) into the marketplace to keep clients from moving to a different non-Microsoft solution.  

The leader in the cloud compute space currently would seem to be Amazon with their EC2/EC3 and S3 Web Services offerings, and while many companies are using these, I have yet to see to any really large companies do so – the length of time it takes to make a decision in a large company typically means that whatever is being looked at is already obsolete long before a deployment is possible. A case in point would be my former employer where, with a bit of luck, Microsoft Vista will finally start to be deployed next year (they are still running Windows 2000) just in time for it to be superseded by Windows 7.

At any rate – there will be lots about this in the press over the next hours, days and weeks so it will be worth keeping an eye on. I personally tend to view cloud computing as a total package – as exemplified by the emergence of net devices (think of the eee PC, and the Android mobile phone) accessing web based document creation, email and media services (like Google Apps which Microsoft is also promising to offer in its next iteration of Office), as well as backend services such as storage, backups, financial applications etc. (as can be sourced on Amazon servers – or XCalibre, or Flexiscale, or Q-Layer, etc. servers) over the air via the internet. After a number of false starts and tentative small steps over the past several years, there does now finally seem to be some industry momentum building.

Putting aside the observation that Microsoft’s key objective is to maintain user lock-in to their product slate by building tight dependancies to other Microsoft technologies, having Microsoft come to the table with their ‘me too’ presentation today is a good thing in the long run for two reasons:

• the competition keeps the industry sharp – this is a rapidly evolving area with winners and losers yet to be decided
• their entry at this juncture adds validity to cloud compute technology putting it squarely on the radar of the larger corporate clients who are already Microsoft shops.

This could start to get quite interesting !

Amazon S3, Backups and Other Things

Most folks these days have a digital camera, and as a result have hundreds and possibly thousands of digital photos and videos littering the hard drives of their computers. Certainly this is true for me and of course it highlights one of the dilemmas if living in a digital age. Crash your hard drive, and your photos are gone forever. Admittedly not all of them are ‘perfect’ shots, but out of every hundred or so there are bound to be a few that are precious to you and which you really do not want to lose.

In prior generations, there were generally lots of ‘hard copies’ of photos either in photo albums or boxes stashed away in closets, and copies of the more memorable photos would be made and sent around to relatives etc. This was a good thing, and back then, if I were to lose a copy of the photo of my great aunt Lulu (I actually don’t have an aunt by that name, but you get the idea) there would have been a good chance I could either dig through my shoe box full of negatives and have a new photo printed, or perhaps could ask a family member for a copy of a photo they had and I’d be back in business. That was then, and this is now.

Like many people, I starting burning CD copies of my photos, and then as the numbers and sizes of the images grew and video clips were added to the mix, eventually DVD copies which I keep in a safety deposit box at a local bank. This has the advantage of it being a relatively cheap option and also of the photos being kept off site so if the house burns down they will still exist. Of course it has the disadvantage of me having to constantly update and replace these discs as new photos are added, which is time consuming and like most people, I’m not quite as disciplined as I should be.

Off site storage for a few photos in something like WebShots is an option which many people use, and there are many others like it, as well as the various ’social network’ solutions such as Facebook, but in general the advertisements drive me crazy, so I refuse to use them for photos. To avoid all the ads, and for displaying a handful of photos that you really like, there is the option of putting a few photos up on your own web site as I have done, or as a friend of mine has done. But what do you do to replace the shoe box full of negatives you used to keep filed away, ‘just in case’, and how do you ensure it is not computer dependant as well as being easy to use ? Well, I’m not sure what you are using, but I am using Amazon S3.

Before retiring from Shell, I did a writeup on the various Amazon Web Service offerings, and I like what they have done – they have platforms for applications development, commercial platforms for managing and running applications, collecting money etc., and they also have a bulk storage solution. The storage solution (called Amazon Simple Storage Service, or S3 for short) distributes your data across many servers, is highly fault tolerant, is accessible from any computer (Linux, Mac or Windows) anywhere that you have a decent internet connection, is hugely scalable, and is pretty cheap (you just pay for what you actually use). All of this makes it quite attractive.

By itself Amazon S3 is not even slightly user friendly, as it is targeted toward developers, but there are a few companies who now offer beta (of course !) software you can use as a user interface for the S3 service to allow you to store all your ’stuff’ on the Amazon S3 servers. I actually have about 17 GB of files sitting up there ‘in the cloud’ and other than a couple of horrible moments which I’ll get to in a minute, I recommend it. For me at least, as a back end storage service Amazon S3 has been stable, reliable and accessible. But you do need a front end interface of some sort to use it, so read on.

Before starting, you need to decide what it is you actually want to be able to do – in this instance, although you can use Amazon S3 as the storage location for the data displayed on your web site, what I was looking for was a replacement for the shoe box I had for my negatives. I admit only a few of the photos are masterpieces (to me at least), but on the other hand saving everything gives me the freedom to pick and choose whenever I want. Having decided on the ’shoe box’ approach, there are quite a number of commercial alpha and beta products out there – many of them are listed in the Amazon Solutions Catalog. Some of these are good – some of these are frankly awful, but all of them demonstrate the possibilities and will get better with time.

Three of the interface solutions that I tried out to allow me to just be able to drag and drop files from my hard drive into Amazon S3 buckets (which is what they call the storage locations you have access to in S3) are S3Fox, Bucket Explorer, and Jungle Disk.

S3Fox looks nice, but I (and many others) could not reliably get it to work – no idea why. For the moment, I’d suggest it be avoided till it has a few of the kinks worked out. It also requires Firefox and does not work with Microsoft Explorer, which I think is OK as Firefox is what I’d recommend anyone to use if asked, but it does freeze them out of a large part of the market

Bucket Explorer and Jungle Disk are roughly similar to the casual eye, but did offer me a few stressful moments. After creating a bucket with Bucket Explorer and filling it with my photos, I then tried to open it with Jungle Disk – no go. Everything appeared to be gone, other the the top most level which only had the bucket name. Likewise when I created a bucket in Jungle Disk, Bucket Explorer could not read the name of the bucket, or the names of the files it contained. I’m sure things like this will get fixed over time, but for the moment – for the casual user – I’d suggest picking one or the other and not to try to mix and match.  By themselves, both work and are good choices.

I will continue looking at new options as they become available, but for anyone looking for a digital shoe box for their digital images, I think the Amazon S3 product with the user interface of your choice is definitely a good option. For the moment I’m happy – and sticking with Bucket Explorer !