One of the more popular arguments made against cloud computing is a perceived lack of useful standards. For example, Dave Linthicum, the CTO and founder of Blue Lab Mountains, mentioned in a recent article:
“…the notion that you can easily move from one provider to another without significant work and cost is largely science fiction at this point.”
While his argument may have a certain degree of technical merit, it still rings hollow. The growth of cloud computing shows no signs of slowing down: major providers display consistently strong growth. Analyst firm Gartner predicts worldwide cloud services spending to surpass $109 billion in 2012 alone. In fact, large enterprises willingly choose multiple clouds, and it’s illuminating to consider the reasons why this happens.
The entire premise of virtualizing your application infrastructure is to give you the ability to divorce your apps from your physical infrastructure on which they are hosted. This, in turn, allows your application workloads to be dynamically placed and migrated across a pool of application server resources, which allows the infrastructure to dynamically adapt and respond to your evolving business needs. If you look at traditional applications and how they are developed, it’s clear they were not designed for the cloud, and they definitely don’t take advantage of some of the best benefits a virtualized infrastructure can offer.
Each of the major cloud environments offers a unique set of benefits and differentiators. Users of AWS don’t choose that platform because they feel compelled to; instead, they choose AWS because it gives them flexibility and services for their particular application requirements. The very same user might select a private VMware-based cloud for a different application because, again, that application has a different set of requirements. Cloud diversity is a good thing because it presents developers a range of choices.
So, yes, when discussing cloud diversity, you can have your cake and eat it, so long as you pick the cloud that is best suited to your:
- Application and services, and it has the right technical capabilities that your application requires
- Business and commercial criteria encompassing the cost imperatives and SLAs you need
- Customer needs, including their geographical proximity, regulatory and data protection laws, etc.
At Riverbed, we see more and more of our customers considering cloud-based architectures as a means to transform their application business models, particularly those with fluctuations in traffic and seasonal demand. Essentially, our customers find that moving to the cloud gives them a competitive advantage, the ability to provide differentiated service offerings, and new revenue models.
Cloud computing isn’t limited to just a collection of virtual machines and storage you rent by the hour in a location far away from your data center. Mature cloud providers offer the ability to extend existing on-premise infrastructures into cloud facilities, creating a unified architecture with the benefits of instant infrastructure. Applications can span both, and users need not notice the difference.
Can I have cloud diversity if part of my app infrastructure is not software?
Here’s the catch. You’ve virtualized your application delivery infrastructure and have started to push some of your apps out into the cloud. But part of your app delivery solution is not software. You have a hardware ADC that is critical to the correct operation of your apps and the vendor provides a virtual appliance. Neither of these are ‘software’ in the sense that they can be deployed anywhere. How is this going to impede and limit your ability to truly virtualize and reap the benefits cloud diversity brings?
When everything is software, including the network and the ADC with robust and open APIs, you get into the realm of a truly programmable infrastructure model. A great way to look at this would be the conventional jet engine of the cloud takes you supersonic, and the scramjet of programmable infrastructure then goes hypersonic.
Yesterday’s load balancers and legacy application delivery controllers are not designed for the cloud and to give you the type of diversity, portability, programmability and granular application-level control. The mismatch is clear.
- Enhancing efficiency and response times of applications and services
- Improving availability between instances that span multiple geographic zones and regions
- Solving latency problems with content optimization and acceleration tools
- Ensuring proper protection using intelligent layer-7 inspection against known and unknown threats
- Scaling resources to provide encryption and compression services without affecting performance.
One example of a software ADC is the Riverbed Stingrayfamily. This new breed of ADC is natively designed for virtualization and cloud portability. As a pure software solution intended for the widest variety of deployments, the Stingray family enables a more flexible application delivery strategy and provides a common delivery and control platform that can grow with your business.
for more information on Riverbed please visit http://www.c24.co.uk