Software-as-a-Service (SaaS), Platform-as-a-Service (PaaS), Infrastructure-as-a-Service (IaaS) – during the “Peak of Inflated Expectations” the promise of reduced OPEX, along with more reliable, flexible and scalable application services made Cloud Computing THE hot technology. Here’s a nice chart that lays out some of the major players and where they play.
Cloud Utilities and Applications
Cloud services, such as Amazon’s Web Services and their Elastic Compute Cloud (EC2), are already providing scalability to Facebook, storage to support Microsoft software downloads and overflow compute capacity for popular real estate website zillow.com. Nevertheless, we seem to have entered the “Trough of Disillusionment” for Cloud Computing.
It remains clear that improved operational competencies, 24 x 365 hardened environments and accumulated economies of scale (facilities, power, compute power, storage and bandwidth) will lead many Enterprises to embrace these new models. Given the current economic backdrop, the need to reduce CAPEX and the need to support IT operations with limited budgets and staff, it’s likely we’ll start hearing about more enterprises migrating application services to cloud computing. Some will be “Private Clouds” in which their Data Centers have been outsourced (enterprises with extreme security & regulatory requirements), while others will begin to embrace the more open service architectures like those offered by Amazon and Google (see picture).
Amazon Web Services vs. Google: Comparing two leading software platforms in the cloud
A major concern that must be addressed to foster this migration is providing enterprise customers with the ability to understand how application performance will be impacted as services are moved into the cloud. How can development and test engineers ensure their applications will prove resilient and meet performance expectations in the unknown cloud environment?
The good news is that cloud services are maturing. Management services are available to requisition machines with desired processing power, memory and storage. Common operations include installing an application environment, managing access permissions and executing applications. Some Service Providers even provide APIs that let enterprises direct application instances in to run in specific geographic locations to add resiliency, or to place applications and data closer to the end user. And with cloud data center locations ranging from San Jose to Sao Paolo, and Beijing to Bangalore, this provides incredible flexibility.
Given this level of control, it is now becoming possible for enterprise QA Engineers and Performance Engineers to stage their Cloud environment in a local test lab. They can model the platforms they will requisition in the cloud, they can use Load Tools to simulate user traffic and they can use WAN Emulation Tools to understand and compare how the applications will perform when hosted at one or more cloud data center locations. Some innovative products, such as Shunra’s VE Network Catcher, can be used to capture network behavior from various end user locations into different cloud data centers so the performance of different scenarios can be understood before you go live.
Have you migrated any applications into a cloud environment? How did you test your applications to ensure they would perform well at all your end user locations?
Leave a Reply