Archive

Delivering on cloud: Do you need a car or a race car?

Share this post:

Car manufacturers make cars. Simple enough, right? 

Thanks to Henry Ford and many successors, car companies can now produce vehicles with incredible efficiency—to the tune of 70 million vehicles per year, worldwide. These companies even support variability in their process for the production of mass customized service vehicles where they manufacture vans as ambulances, pickup trucks as tow trucks or cars as police cars or taxis.

In many ways, car manufacturers are like infrastructure as a service (IaaS) cloud services, and I’ll explore this more in a bit.

But a car alone is not going to win the Daytona 500. 

Even if the car manufacturer paints it a bright color, puts on wider tires and installs a bigger engine, it’s still not a race car.  Without a driver, sponsors, a racing license, a pit crew and so on, it’s just a car in a fancy costume and won’t win any race with serious competition.

No matter how fast the car manufacturing company produces that car, or how many cars they can produce at once, it may not improve the time it takes for a race team to get that car onto the track.  If the critical path is getting the racing license, or hiring and training a driver, then reducing the time for delivery of the new car (or 50, or 1000) won’t really buy them much.

The same is often true for servers provisioned by IaaS.

IaaS providers can now spit out virtual machines (VMs) at breakneck speed and volume. SoftLayer takes just minutes to deploy customized bare metal servers, but if the end use for those servers is not taken into account then there may be many more slow and costly steps required to make that server finally useable for business.

The companies I work with pay IBM to manage servers for them, among other things. We have to pay penalties if anything breaks, fails an audit or gets hacked.  So, in order to fulfill expectations we will have to perform some combination of these following services (and more) on somewhat of an ongoing basis:

  • Asset and configuration management
  • Backup and restore
  • Change management
  • Health checking and patching
  • ID management
  • Vulnerability scanning

For more on IBM Managed Services, please watch this video.

We could certainly debate where transforming customer expectations for hosting fits in here, but let’s leave that for future blog posts and for now just recognize that some customers want us to manage cloud servers in the way we manage their non-cloud virtual machines (VMs).  They want to dip their toes into the cloud with expectations of speed or savings, but sometimes forget to look at the big picture.

They want a race car, but IaaS only gives them a car. 

Once an IaaS service provisions a server they might still require many additional steps for activation, some of which might take much longer than the provisioning of the bare-bones VM.   Installing an agent on the VM, registering the VM endpoint with a central server (or service), using the agent on the endpoint for the first time or gathering evidence that it all worked properly may become necessary.

And even if they get the car quickly, it’s still not a race car.

Now, we could begin to debate what service management tool and process transformation is required to improve speed or cost in a cloud computing context, and I promise to handle those in future blog posts. This is a critically important topic because if your IaaS is fast and cheap but ongoing server management works like a tired old sloth, then cloud may never let you realize the speed or cost gains you desire.

But for now it’s just important to understand that when you provision managed servers, you have to do a lot more than basic IaaS providers who spit out bare-bones VMs, mail you the root password and wish you good luck with your unmanaged server.

Try calling that provider to lay blame when the VM gets hacked.

Good luck with that!

So, sometimes a car in a fancy costume just isn’t enough…

Make sure you think hard about whether unmanaged VMs are really sufficient for your business needs.  If you aren’t sure, please comment on this post or follow me on Twitter @Stewart_Hyman, and I’ll be glad to help.

More stories

Why we added new map tools to Netcool

I had the opportunity to visit a number of telecommunications clients using IBM Netcool over the last year. We frequently discussed the benefits of have a geographically mapped view of topology. Not just because it was nice “eye candy” in the Network Operations Center (NOC), but because it gives an important geographically-based view of network […]

Continue reading

How to streamline continuous delivery through better auditing

IT managers, does this sound familiar? Just when everything is running smoothly, you encounter the release management process in place for upgrading business applications in the production environment. You get an error notification in one of the workflows running the release management process. It can be especially frustrating when the error is coming from the […]

Continue reading

Want to see the latest from WebSphere Liberty? Join our webcast

We just released the latest release of WebSphere Liberty, 16.0.0.4. It includes many new enhancements to its security, database management and overall performance. Interested in what’s new? Join our webcast on January 11, 2017. Why? Read on. I used to take time to reflect on the year behind me as the calendar year closed out, […]

Continue reading