Automated Scaling Listener

from Wikipedia, the free encyclopedia

An Automated Scaling listener ( German  automatic scaling monitor ) is in the cloud computing region of the computing equipment, a service , the communication and workload from other services via network and application monitoring monitored to autoscaling allow.

functionality

In the event of an (imminent) overload or failure of a service, the Automated Scaling Listener creates additional instances of the corresponding service ( horizontal scaling ) or moves it to a more powerful computer ( vertical scaling ). If predefined limits are exceeded, the Automated Scaling Listener can refrain from further instances and notify an administrator .

Classic Automated Scaling Listeners work here on a rule-based basis and generate additional instances when a certain percentage of the instances of a service is utilized above a threshold value. However, since the creation of additional instances takes a lot of time and a reserve must always be provided, which requires a lot of energy and computing capacity, automated scaling listeners are also used, which use machine learning to make a prediction about the expected workload.

swell

  1. ↑ a b Automated Scaling Listener. In: Cloud Patterns. Arcitura Education Inc. , accessed May 7, 2017 .
  2. ↑ a b Autoscaling. In: Microsoft Azure Documentation. Microsoft, July 13, 2017, accessed May 8, 2017 .
  3. ↑ James Vincent: Google uses DeepMind AI to cut data center energy bills. The AI ​​successfully reduced power consumption by 15 percent overall. July 21, 2016, accessed May 8, 2017 .