***Cloud Services Management for Network Function Virtualization***
**Main Objective** : To provide a reliable and efficient service to the client by creating a smart cloud platform.
* Sub Objective 1: To monitor the traffic within the cloud network, hence make decisions accordingly to improve the availability of resources throughout the network.
*
* Sub Objective 2: To distribute the workload across multiple computing devices, thus maintaining a suitable network traffic load balancing strategy throughout the cloud.
*
* Sub Objective 3: To predict the traffic within the network which will directly contribute in making decisions about future modifications and addons to the system.
*
* Sub Objective 4: To allocate the resources according to the client’s need , through utilization and optimization of resources present.
***Cloud Services Management for Network Function Virtualization***
**Main Research Question** : Is there a platform for cloud services providers with NFV which provide
Network traffic monitoring
Load balancing
Future predictions
Resource allocations
using a single platform ?
**Main Objective** : To provide a reliable and efficient service to the client by creating a smart cloud platform.
* Sub Objective 1: To monitor the traffic within the cloud network, hence make decisions accordingly to improve the availability of resources throughout the network.
* Sub Objective 2: To distribute the workload across multiple computing devices, thus maintaining a suitable network traffic load balancing strategy throughout the cloud.
* Sub Objective 3: To predict the traffic within the network which will directly contribute in making decisions about future modifications and addons to the system.
* Sub Objective 4: To allocate the resources according to the client’s need , through utilization and optimization of resources present.
There are few ways that we can conduct the cloud monitoring. They are virtual network monitoring, website monitoring, cloud storage and database monitoring, and virtual machine monitoring. We can monitor the network traffic through 24/7 or periodically using algorithms. The selected strategy must be able to identify when service-level objectives (SLOs) are not being met, likely negatively affecting the user experience. The cloud administrator can review the operational status and health of any cloud-based device or component using the results obtained with the implementation of monitoring strategy.
Load balancing in software defined network virtualization can be achieved at two levels: NFV level and SDN level. In order to achieve a load balancing in the SDN network, multiple selections for the least loaded/busy real-time server (RLSs) are made. The task of RLS is to choose the most appropriate route for forwarding each new stream and it is based on network time statuses. Algorithm or routing protocol should be implemented in the routable devices to identify the best path. Also, there are static and dynamic load balancing algorithms.
Network traffic prediction aims at predicting the subsequent network traffic by using the previous network traffic data. It will be a proactive approach for network management and planning tasks accordingly. In this we will be using predictors such as time series-based predictors, artificial neural networks-based predictors, and wavelet transform-based predictors.
*Member 4 : IT18099928 - Resource allocation*
There are several resource allocation strategies (RAS) to use according to the client’s requirement and fulfill the needs of the client. For an example, Execution Time - Based RAS, VM-Based RAS and Reservation Based RAS can be considered. The strategy will be selected considering the following scenarios.
• Under provisioning or over provisioning of resources
