CIOREVIEW >> Cloud >>

Auto Scaling In Rackspace

By CIOReview | Tuesday, March 15, 2016

Fremont, CA: Web infrastructure hosting is an emerging trend in cloud technology. While there are many hosting agencies in the market, Rackspace differentiates itself with its easy and intuitive interface to set up servers and load balancers. Further, the firm also offers auto scaling functionality in its data centers spanned across four continents.

Generally, auto scaling is offered as ‘schedule scaling’ and it requires a pre-specified time at which specific number of servers will spin up or down. It is useful to organizations that can predict accurately the rate and time of web traffic on their websites. But when it comes to real time traffic analysis, the pattern is more or less random, giving less space for precise forecasting. In this scenario, LiquiD AutoScaler is emerging as an external measurement platform that effectively communicates with Rackspace's API utilizing scripts. Subsequently, the platform instructs Rackspace to scale resources based real time web traffic, describes Mart Wesseling.

For companies experiencing fluctuating traffic patterns over a 24 hour interval, implementing flexible deployment of servers is the most probable solution. It helps in preventing websites from becoming slow, or even going offline. Liquid AutoScaler analyzes the performance of load balancers to monitor the website under consideration. In the event of slow page loads, it instructs Rackspace to spin extra virtual servers and attach them to load balancer. The time taken to launch new servers and join them with load balancer’s worker pool is that of a few seconds and depends upon the provisioning queue of Rackspace’s cloud environment. Further, LiquiD AutoScaler takes about five minutes for initial set up.

Liquid AutoScaler significantly cuts web infrastructure running costs by using servers only when needed. It positively impacts environment by saving energy on servers that are not in use. It is also able to make real-time scaling decisions without hooking up to third-party data sources. Keeping end user’s experience at the heart, it ensures that users experience same regardless of the volume of the traffic.