Six rules for evaluating cloud computing service providers

In hosting mission-critical applications, companies are starting to rely on cloud vendors, which increases the risk of choosing the right cloud computing services. So how should enterprises make the right decision in such a dazzling mess? Here are some key factors for evaluating cloud computing services and resources available for use.

Performance

One of the concerns that companies consider cloud computing is performance. Implementing an application's high-speed delivery in the cloud is a challenging proposition that involves many factors, including a holistic approach to access and an end-to-end view of the application's request-response path.

Performance issues include the level of I/O access between the application and the data relative to the end user, the cloud, inside and outside the cloud, and between the computer layer and the data store. Recently, numerous services such as Cloudsleuth and Cloudharmony http://www.aliyun.com/zixun/aggregation/8503.html "> The study attempts to measure and evaluate the performance of the services provided by cloud computing providers from a variety of geographic locations and different application approaches.

Technology stack

For a long time, some cloud computing providers are dedicated to providing services for specific software stacks. Typically, this means that they shift from an infrastructure, a service (IaaS), to a platform, a service (PaaS). Different stack-specific clouds can match most of their popular software stacks.

Typical representatives include: Heroku and Engineyard Ruby; Vmforce and Google Application engine (Gae) java/spring (where Gae also supports Python), Phpfog php and Microsoft Windowsazure. NET.

If your application builds with one of these stacks, you might want to consider these cloud computing platforms. They can help you with the installation and configuration of low-level infrastructure, saving you a huge amount of time and expense. On the other hand, they tend to require developers to follow certain recommended requirements when architecting and writing applications to facilitate the creation of high-level vendor-level synchronization.

Service-level agreements and reliability

Some cloud vendors have made a commitment to provide higher levels of service, in a gesture that is different from competitors in other industries. For example, Rackspace provides a higher level cloud service SLA, which competes with giant Amazon, the cloud computing industry.

Note that SLAs are usually only one indication of the result when a service fails, not the reliability of the actual service. The typical representative is gogrid10000% guaranteed SLAs. In other words, Gogrid made a commitment to 100% uptime. If it fails to meet this level of reliability, it will compensate the user for 100 times times the cost of downtime.

While SLAs are an effective indicator of all vendor commitment levels, it is a more difficult challenge to understand the actual uptime of a particular cloud vendor. Most vendors provide a status page to indicate that the service is functioning correctly, but these pages usually display data only a few days ago or even earlier. To achieve long-term data that is actually reliable and available, it is best for users to rely on customer feedback and comparison services such as Cloudsleuth and Cloudharmony.

(Responsible editor: The good of the Legacy)