7 AUTO-SCALING COMMUNICATION FRAMEWORK When the load of multi-stage service changes, the number of GPUs/n- odes changes in consequence and the microservices are also dynam- ically deployed. For intra-node communications between microser- vice instances, servers (such as the DGX-1) have both ...
evaluating its performance using realistic workloads and works suggested in the state of the art. CPU utilization and request rate are mainly used to estimate the workload of an application and guide autoscaling. Chameleon combines forecasting methods and realtime monitoring to enable proactive and re...
Function as a Service (FaaS) provides enterprises with a cloud-native serverless solution to build robust, scalable, and loosely-coupled distributed applications with a low operational cost. Clients can use such a platform to encapsulate the complex business logic into independent micro-services that ...
The ARIMA (autoregressive integrated moving average model) state space model is a time prediction model that combines autoregressive (auto regressive, AR) and moving average (moving average, MA) models. The ARIMA model regards a time-based object sequence as a random sequence. According to the au...
For the complete satisfaction of the client, execution of tasks should be as per the QoS parameters; hence a QoS aware cloud framework is required for the purpose mapping of resources efficiently. To handle the complex issue of the resource placement problem, a cloud architectural framework named...