• VMware announces VMware vCloud Express, goes head to head with Amazon EC2
  • VMware today announced vCloud Express, a new class of service that will deliver on-demand, pay-as-you-go computing power as a service, much like Amazon Web Services’ Elastic Compute Cloud (EC2).
• A PCI-Compliant Cloud? Not at Amazon
  • Very interesting debate and conclusion regarding Amazon EC2 & PCI. Worth a read!
• Azure Reference Architecture Explored with Project Riviera
  • If you’ve been wanting to see how a multi-tenant architecture works with Windows Azure Platform, you’ll want to see Project Riviera. The project has been released on MSDN and includes source code. Interestingly the sample application is a loyalty application!
• Cloud Cartography & Side Channel Attacks
  • This is obviously dependent on the availability of a weakness in the hypervisor, but definitely an interesting concept none the less.

Security

• Microsoft Internet Information Server (IIS) FTP Server Buffer Overflow Lets Remote Authenticated Users Execute Arbitrary Code
  • Any one running IIS 5, 5.1, or 6 and FTP – watch out!
• New version of WiKID authentication server
  • WiKID Systems announced version 3.4 of the WiKID Strong Authentication Server in Enterprise and Community Editions. New features include built-in support for the SAML Single Sign-On Service for Google Apps, a new self-registration process for Active Directory users and extended vendor-specific RADIUS attributes.
  • Pretty interesting stuff, not seen this before.

Hopefully my writers block has been solved and I’ll get back to writing some code

Gareth

Definitely pretty cool stuff. Now to see how well it performs and verify it supports all the nice SQ L stuff  (sparse columns, PIVOT, stored procedures etc).

This will certainly be hyped by Microsoft over the coming weeks, and we will be trying it out! Much more to come on this topic,

Gareth

Microsoft is updating the branding for SQL Services and SQL Data Services. Effective immediately, SQL Services will be called Microsoft SQL Azure, and SQL Data Services will be Microsoft SQL Azure Database.

There will be more news and updates related to our Software + Services strategy next week at Microsoft’s Worldwide Partner Conference in New Orleans. For those of you not attending in person, all the news and information can be found at: http://www.digitalwpc.com/ . Check back often starting next Monday, July 13, for daily updates.

More about TSQL Support in SQL Data Services can be found here. Another announcement of this can be found at “SQL Azure – Same great platform, just a better aligned name”

Apparently its a ‘Cloudy’ week

Gareth

• Amazon Adds CloudWatch Monitoring, Other Services
• Amazon to ‘Open Source’ their cloud API’s
  • This will be a very telling move if it gains acceptance. It will open up the way for a number of people to standardize to, but I am more than a little skeptical
• Credit Card Council Looks Into Cloud Security
  • You knew it had to be coming – and now it has!
• Bulk Loading/Export from Clouds
  • This shows how the systems are starting to mature. It has always been a fear how to get large data sets to and from cloud systems. Specifically I imagine there will be data analysis cloud applications that may end up processing terabytes (or more) of data. Getting that information there was a problem until now.
• Cloud Electronic Socket for $100
  • This is key, this moves backup from significant chunky machines (PC’s) to a small appliance no larger than an electrical adapter. Very nice, and I hope successful.
  • Socket appliance backs up to cloud storage
• Tibco enters the cloud space using Silver on Amazon
• Microsoft and Sun talk interoperability

The clouds seem to fall into the following categories:

• Amazon Elastic Compute Cloud –
• Microsoft Azure
• Google Apps running on the App Engine
• IBM – Cloud running on platform provided by the new Cloud division
• Open provider – Basically a OSS company aiming to break into the emerging market

Outside of the Open provider I’m afraid that these are basically going to be our cloud choices. Purely based off the fact that the benefit of a cloud means that the cloud provider has to have a heck of a lot of processing infrastructure setup. Its really not a world where you can deliver your solution from your garage after from 2am hacking through the code base. We have moved from the inventive ‘anything goes’ era, to the commodity era. Even with VC backing it would be very hard to become a Open provider of cloud resources – I hope I’m wrong, but I dont believe any VC will stump up the vast amount of cash to get into the party this late into the game.

So what cloud offerings are available to us who use C# (CSharpee’s)? I dont see Google promoting C#, their languages of choice seem to be JavaScript and Python (although I see Java starting too wiggle in). So thats 2 out (Google and the Open provider).

I’m not sure I truly follow IBM’s cloud strategy at this time. Rather than providing the infrastructure they appear happy to provide the software to run within the Amazon cloud. So unless they are looking to provide some private cloud infrastructure they appear to be happy to just run within other companies clouds. If that is the case then for this part of the discussion as a cloud vendor they are out as well.

So that leaves us Microsoft and Amazon. Now things definitely get interesting, since Microsoft own the Microsoft platform they are probably (anyone hear anti-trust court hearing coming…) not ‘unofficially’ bound by the same pricing issues as Amazon using the Microsoft platform. Fundamentally Amazon charges 25% more to use a Microsoft platform over a Linux platform. So anyone looking to run a C# application must be reconsidering the language choice if they are hoping to need many servers running a successful software solution , naturally Microsoft pricing should (and I say should as at the time of writing it hasn’t been released) be competitive.

Doom & Gloom? I don’t think so, if you love C# (like me) I don’t think its as bad as the picture above paints. The savior is something we have all heard about, but not necessarily thought about in this way. The Mono Project actually provides the answer to any cloud cost based dilemmas. For the longest time I knew about Mono, but couldn’t reconcile in my head the compelling reason (any reason actually ) why I would ever have the need to use it. I work in a Microsoft shop, all our tools are Microsoft based (although that is changing with the ‘appliance’ concept) – why would I need my C# to run on anything other than a Microsoft platform? Now Clouds have brought in the compelling reasons – price and scalability.

I’m intentionally skipping past the fact that Azure is positioned to be more than the Amazon ‘virtualization’ cloud and very well beat it in price due to its different approach. However the thing Microsoft cant avoid is running on Linux is ‘on the surface’ cheaper, so any Azure pricing has to beat C# running on a Linux platform. Which really means that as long as Mono is viable you still have a hosting choice where you can run with the same, or hopefully better if Azure is what I think it will be, as the Linux hosted/clouded systems. Predominately because it can run on those systems .

Mono is certainly interesting, and while it still needs some more hardening (see a upcoming post on Mono garbage collection) I imagine the Cloud race and Mono positioning will make them become the best of friends.

If you are thinking about putting some stuff up in a cloud and where waiting for Azure, perhaps you should have a look at Mono. You may be pleasantly surprised.

Gareth

# Guest Failover Clustering is supported for SQL Server 2005 and SQL Server 2008 in a virtual machine for Windows Server 2008 with Hyper-V, Microsoft Hyper-V Server 2008, and SVVP certified configurations provided both of the following requirements are met:

* The Operating System running in the virtual machine (the “Guest Operating System”) is Windows Server 2008 or higher
* The virtualization environment meets the requirements of Windows 2008 Failover Clustering as documented at The Microsoft Support Policy for Windows Server 2008 Failover Clusters.

Guest Failover Clustering is when you create a SQL Server failover cluster inside a virtual machine where the nodes are running as a virtual machine. A non-SVVP configuration that meets these requirements will receive support from Microsoft CSS per the policies documented in

897615 (http://support.microsoft.com/kb/897615/) Support policy for Microsoft software running in non-Microsoft hardware virtualization software

Very cool, the Data Center guys are going to love this one!

Gareth

* Server 2008 Hyper-V 16 LP Support
* Server 2008 Hyper-V +update (KB95670) 24 LP Support
* Server 2008 R2 Hyper-V Original POR 32 LP Support
* Server 2008 R2 Hyper-V RC/RTM 64 LP Support!

Stunning difference! It definitely seems VMWare has awoken the slumbering giant. This in conjunction with Live Migration – I have to say things are starting to look good as a developer , roll on Windows 2008 RC2!

I have to say I’m really looking forward to this release, I think IT shops will start to push developers into smaller scalable instances of the VMs. Exactly (as it appears) Microsoft is doing with the embedded XP mode within Windows 7.

And don’t forget that R2 is a pure 64 bit OS, so this is really a milestone release – and it could be argued R2 is an understatement!

Gareth

So in rebuttal to this the SQL team issued the Windows2008 R2 beta download runs smoothly now blog. Ah yes you can imagine the response from the SQL team when they heard about this, talk about high profile egg! I can imagine the conversations “You did what? Why didnt you talk to us first…”, I removed the bleeps

That all being said I suspect this will be one of the last times we see this happening. There will be no doubt lots of flying stuff going on in MS regarding this, but I suspect that this potential embarrassment will make MS start to work towards a scalable ‘cloud’ set of resources. One that can quickly scale to demand and drop back down again – sound familiar ? The SQL part of Azure (well cloud computing) should take care of this particular glitch once and for all, and I suspect this incident will drive this direction for Microsoft internally. People building their own SQL infrastructure will be dropping in favor of scalable cloud resources.

Adding to that mix the Register is reporting that Oracle runs better on VMWare compared to their own RAC solution. Which is really an early ‘cloud’ like offering. It seems clouds are here to stay – if only for availability and scalability abilities.

Ah yes interesting times!

Gareth

I suspect in the near future gone are the days of setting up expensive OS (Windows or otherwise) clusters, and move in dedicated cluster appliances. Each running a set of host ‘machines’ that can ‘transparently’ fail over. It seems that as hard as the various OS vendors have tried, it appears the VM approach seems to have won over the fickle IT crowd.

As we would expect it wasn’t necessarily the technical skills that wows the crowds, but really the $. Enterprise versions of the various OS’s often cost more in licenses and more to maintain. In addition they are more complicated than managing a single machine. So what is the downside of managing a highly available single machine?

Has anyone tried to image and restore a physical clustered machine? It certainly wasnt trivial when some friends of mine tried. VM’s are so nice and easy to deal with – “Need to add a drive? Ok then I’ll allocate one for you”.

Roll on Simplicity is what I say!

Gareth

Google’s note on Cloud computing:

http://googleenterprise.blogspot.com/2009/04/what-we-talk-about-when-we-talk-about.html

ZDNet comments on ‘private clouds’:

http://blogs.zdnet.com/microsoft/?p=2610

These are interesting as the real benefit to this concept is that there is a consistent API that would allow applications running in a private cloud to be moved to a public cloud. Without this I don’t see how private clouds will really survive as a long term concept. Basically will the commercial cloud players try and make a stab at a uniform interface (or collection of interfaces)? If that happens the private cloud concept will live and prosper, otherwise it will take a couple of years to shake out the players until there is a defacto set of standards.

Finally The register has:

http://www.theregister.co.uk/2009/04/29/eucalyptus_goes_commerical/

This is where a newly born company (out of academia) is duplicating the Amazon Web Services (AWS) and enabling the private cloud concept. Pretty humorous amount of activity yesterday

Things are definitely hotting up under the clouds

Gareth

Firstly I don’t think the phenomenon it really restricted to technology, its really the coming together of a number of factors that truly make this concept pivotal to IT groups.

• Scaling technology to this type of demand has matured to the point this is not longer a pipe dream
• IT is becoming more and more commoditized and is really merging with the business community
• Business is getting much more savvy with IT technology
• When IT gets in the way, the business will find a way around it

So before we start with the factors, what is a cloud? Fundamentally cloud computing is having a virtual resource available in a remote network. In theory it is infinitely scalable and easy to setup. Everyone knows about virtualization where the VM machines are running on a VM host (on their own machine or in the data center), a version of a cloud (Amazon for example) is really running virtual machines in a virtual remote cloud. You don’t need to get IT to allocation you hardware in the data center, as it doesn’t require any! The cost is money but the time to market should be pretty darn fast.

So lets get back to the list.

Scaling technology

If some one had told me 15+ years ago Google would have indexed and cached however many web pages they do cache, I probably wouldn’t have believed them. We are in a position today that few people could have imagined 10 years ago, fundamentally the cost and pure computing power has been linked together to drive cost per work unit way down and availability right up there. I applaud both the search engine teams and the VM companies for driving the innovation in these areas. We we now in a position where we can have virtual appliances out in the web that ‘really’ do exist and the framework to support them allows these to be replicated around the world.

IT Commodization

When IT started it was a highly specialist sector. The business tolerated the IT folks, on the grounds that they knew they needed them if they were going to stay competitive in the market. However if the business could avoid the IT processes it would no doubt try , especially if it would help speed up their time to market. IT groups have long been an aligned arm of the business where there was an uneasy relationship between the two. Now a lot of the specialism that required highly skilled individuals has been moved over to the computers them selves and the load on the IT specialists has changed some – but fundamentally the required skill sets have largely been lowered. I cringe when I write generalizations like that – but it is just that, in general this is true. In combination to these smarter more manageable systems, networks have become so good that supporting a system out of another country is nothing news worthy, in fact its often news worthy when companies DONT offshore. It is now normal to be able to shop around for your IT skills, and the business is not restricted to having the need to keep people local. IT development is going to get merged with the business arm, no question about it – and fundamentally makes sense!

Business is getting more IT savvy

Gone are the days that IT could say – no way “you” can do that. The focus of most tools sets these days are to empower the business community to feed them selves ( See “Give Business Users Control” http://www.intelligententerprise.com/showArticle.jhtml?articleID=217200041 as a recent example). Now days its not uncommon for the business users to be able to generate their own data analysis cubes and lever the data mining objects to perform very sophisticated analysis without needing coders, DBA’s or IT analysts. Since the tools are targeting to the business, the goal is to make it as easy as possible for them to get started. The problem for the business is that while it is easy it doesn’t necessarily mean it scales.

If IT gets in the way, the business avoids it

This one is the kicker. IT is often beaten by the business for not doing x,y,z (and sometimes appropriately – but I wont go there for this article ) . As such IT is often forced, coerced or suggested that they use more process to help avoid any previous issues (aka ‘opportunities’) . All sounds good, the problem is while process is inherently a good thing, it often adds in project lags (due to following the process). Sometimes to the point where if the project was done the value of the project is minimized to such a point because of the amount of time it takes to come to market. This is obviously not business friendly! So with the advent of the cloud computing the business is in the position to get a virtual ‘appliance’, and possibly start to use this without even involving IT groups. None of the normal safeguards are really in place any more.

I suspect the advent of the cloud architecture will be something similar to the RAD revolution that happened over a decade ago. All good stuff, but it needs to be managed well – or long term sustainability of that specific project will be jeopardized.

But I think the whole cloud concept is a winner, more by nature of the conditions and timing than pure technology, but its here to stay.