Comments
Hidden Cloud Costs: AWS, Azure Management Costs Compared
Threaded  |  Newest First  |  Oldest First
Laurianne
50%
50%
Laurianne,
User Rank: Author
8/14/2014 | 1:25:42 PM
good context
Good context on the Red Hat Linux issues Charlie. Readers, how has this issue shaped your thinking?
Charlie Babcock
50%
50%
Charlie Babcock,
User Rank: Author
8/14/2014 | 4:00:23 PM
Azure, Amazon and what about VMware?
As you review management costs for Azure and Amazon, it's hard not to see how VMware has a shot at grabbing part of the bybrid cloud market. The pretty good things that Windows Server, Hyper-V and System Center can do will be amplified by all the things that vSphere and vCloud can do. Think hybrid computing based on live migration of virtual machines. That's not possible today. The same storage file system must underlie both departure point and host destination for live migration. But I'll bet VMware is working on it.
Stratustician
50%
50%
Stratustician,
User Rank: Ninja
8/14/2014 | 6:39:02 PM
Just a wee bit of fine print...
Is it only me who finds "If an AWS server hardware dies and your application goes with it, that's your tough luck. It's up to you to monitor CloudWatch or set up alerts to let you know when a virtual machine has died." really worrisome?

Companies are trusting AWS with their data and applications, and if the hardware, not the virtual machine, but the hardware dies, it's the customer's responsibility to plan for it? Does this not seem a little odd?
Thomas Claburn
50%
50%
Thomas Claburn,
User Rank: Author
8/14/2014 | 6:55:43 PM
Re: Just a wee bit of fine print...
That's news to me. I had always assumed AWS provided enough redundancy and failover that the risk of a server going out wasn't an issue.
PCComf
50%
50%
PCComf,
User Rank: Apprentice
8/15/2014 | 1:19:01 PM
Re: Just a wee bit of fine print...
You still need application level monitoring to ensure that what you think you are providing is actually what you are providing. This is true in and out of the cloud.

I have a very small AWS environment, but I once saw a server fail, yet it appeared to still be running fine in the aws administration console. It took an hour or so to respond enough that I could even manage it enough to restart it / shut it down.

If possible, something outside of your cloud environment should be checking to make sure that services are still up and available rather than relying on built-in cloud tools to keep you going. There are too many ways for a server to "go out" that no cloud provider can account for all of them.
cooljames
50%
50%
cooljames,
User Rank: Apprentice
8/15/2014 | 3:10:25 PM
Re: Just a wee bit of fine print...
That angle is deceptive, though I'm not alleging it as necessarily intentional. When you have auto-scaling turned on, a dead or underperforming (by virtue of your customer config) instance is replaced by another instance. Even if you're running 1 instance, you should turn on auto-scaling and set its limit to 1 if you want this limited notion of availability.

What the article says is that the same instance isn't rebooted. That's a good, smart thing. Since the system can't determine whether the reason for the dead instance falls on the software or on the instance itself, the design of the system covers all the cases. Terminating the failed instance and starting a 2nd is the smartest thing to do there. Also, specifically, opting to restart the failed instance is both proven to be risky and reflective of other cloud provders' limited experience working with the kinds of products that need cloud hosting.

AWS is Amazon's own infrastructure. Their huge success was built entirely on this architecture. Other companies like Rackspace and Microsoft or don't have that same level of insight into operational excellence with regard to online reliability. When you see an article that views AWS in a critical light compared to other cloud hosting products, take it with a grain of salt (or suspect that it's written by an Azure consultant, not the case here).
Charlie Babcock
50%
50%
Charlie Babcock,
User Rank: Author
8/15/2014 | 6:16:46 PM
Watch your app from outside the cloud
"If possible, something outside of your cloud environment should be checking to make sure that services are still up and available rather than relying on built-in cloud tools." Exactly, PCComf. There's more than a cottage industry trying to provide this: think Compuware, Solarwinds, New Relic, AppDynamic, NetScount and don't forget the French firm Cedexis. All can supply info. that the cloud provider may not.
DougL231
50%
50%
DougL231,
User Rank: Apprentice
8/18/2014 | 9:35:02 AM
What about PaaS?
This is useful information as long as you have a large investement in on-prem.  However, it seems to completely ignore PaaS which scales very differently.  If you just want to move your datacenter off prem, you aren't going to see as much ROI as you would if you reduced your overall footprint.  You will achieve some economies of scale and might get some reduced licensing (i.e. you can run SQL on a VM using an Azure per-minute license instead of paying per core), but your cloud provider will be adding profit that otherwise wouldn't be in the mix.  In my opinion, the biggest could ROI is in eliminating servers and environments that exist to handle periodic loads (quarterly finanicals, annual MRP runs, etc.) and otherwise sit unused.

The statement, "Azure's Autoscale allows customers only to turn VMs on and off," is a little misleading in my opinion because that makes it sound like your only option to scale in Azure is to add more VMs and more instances.  That simply isn't the case.  Services such as SQL server can scale without adding more instances.  More importantly, with a PaaS model, you can treat the compute resources as disposable assets and only have to pay for them when they are used.

For a small business or startup, I don't know why you would design VMs into your strategy when you could most likely use a combination of SaaS and PaaS to meet your needs.  For larger compaines, obviously there is an investment required to deploy a work load to a PaaS environment when you already have everything setup to run on your on-prem servers.  However if your cloud strategy is to just take existing server instances and move them to the cloud, I'm not surprised if you are scrambling to find the ROI.
Charlie Babcock
50%
50%
Charlie Babcock,
User Rank: Author
8/18/2014 | 2:10:47 PM
My cost of managemenet comments reflect hybrid-cloud assumptions
Yes, DougL321, my comments reflect an assumption that someone wants a hybrid cloud operation, with some operations on-premises and some workloads in the cloud. Picture looks different for a startup with no on-premises data center; and right, it's also different for someone only interested in platform as a service. How many companies are only interested in PaaS?
DevanshC285
50%
50%
DevanshC285,
User Rank: Apprentice
9/4/2014 | 5:55:57 PM
Offload the Nitty Gritty to partners in Crime- Focus on Strategy!
The Cloudvertical site does offer comparisons to cloud providers' costs. Although location also would play a role. It is also important to avoid missteps in the beginning of this journey as well as avoid the hidden costs down the line.

     Every major IT vendor e.g Google, IBM, Hewlett-Packard, Oracle, Vmware etc -- and the telecommunication service providers who have the Infra resources to run ITaaS offer enterprise public cloud services and want to expand their footprints make revenue off of the cloud. Many of them, notably Rackspace, and HP, are betting on OpenStack infrastructures, which besides Azure, is the most formidable competitor to Amazon.

    It's important not to lose time and effort -Despite the tit-for-tat price cuts in the cloud,  one needs to justify the bigger gains with best practices, as well as avoid those hidden bear traps.  Having a partner that will understand your business requirements, development strategies, outlook, end goal, PaaS/IaaS needs and complimenting the knowledge of the public/private- HYBRID cloud be it AWS or Azure will surely benefit your roadmap by putting you on the right track from the get going.

    As part of the Mindtree's Infrastructure Managed Services team, we have a history of driving/navigating the cloud-wagon for our customers and partners alike as their trusted advisors, while maintain our Vendor agnostic best practices and domain expertise to support your/their needs, implementation and journey.


The Business of Going Digital
The Business of Going Digital
Digital business isn't about changing code; it's about changing what legacy sales, distribution, customer service, and product groups do in the new digital age. It's about bringing big data analytics, mobile, social, marketing automation, cloud computing, and the app economy together to launch new products and services. We're seeing new titles in this digital revolution, new responsibilities, new business models, and major shifts in technology spending.
Register for InformationWeek Newsletters
White Papers
Current Issue
InformationWeek Tech Digest Septermber 14, 2014
It doesn't matter whether your e-commerce D-Day is Black Friday, tax day, or some random Thursday when a post goes viral. Your websites need to be ready.
Flash Poll
Video
Slideshows
Twitter Feed
InformationWeek Radio
Sponsored Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.