Adaptive computing resource allocation approach for virtual network functions
The present invention provides an adaptive computing resource allocation approach for virtual network functions, including the following two steps: Step 1: predicting VNFs' real-time computing resource requirements; Step 1.1: offline profiling different types of VNFs, to obtain a parameter relation between the required amount of computing resources and the ingress packet rate; Step 1.2: online monitoring the network traffic information of each VNF, and predicting VNFs' required amount of computing resources with combination of the parameters in Step 1.1; Step 2: reallocating computing resources based on VNFs' resource requirements. The computing resource allocation approach includes a direct allocation approach and an incremental approach. The adaptive computing resource allocation approach for virtual network functions of the present invention allocates computing resources based on VNFs' actual requirements, and remedies performance bottlenecks caused by fair allocation.
1. An adaptive computing resource allocation method for virtual network functions, the method comprising at least two steps as follows:
step 1: predicting real-time computing resource requirements of the virtual network functions; and
step 2: reallocating computing resources based on the real-time computing resource requirements of the virtual network functions;
wherein step 1 comprises:
step 1.1: offline profiling different types of the virtual network functions, to obtain a relationship between a required amount of the computing resources and an ingress packet rate parameter; and
step 1.2: online monitoring network traffic information of each of the virtual network functions, and predicting the required amount of the computing resources of the virtual network functions with a combination of the ingress packet rate parameter in step 1.1.
2. The adaptive computing resource allocation method for virtual network functions according to claim 1 , wherein the adaptive computing resource allocation method comprises a direct allocation method and an incremental method.
3. The adaptive computing resource allocation method for virtual network functions according to claim 2 , wherein steps of step 2 adopting the incremental method comprise:
step 2.1: counting an actually used amount of the computing resources of each of the virtual network functions, then calculating a difference value between the actually used amount of the computing resources and the required amount of the computing resources of each of the virtual network functions, to find out a virtual network function with a smallest difference value, and;
step 2.2: modifying a priority of a virtual central processing unit (vCPU) thread corresponding to the virtual network function with the smallest difference value, and increasing a supplied amount of the computing resources to the virtual network function with the smallest difference value to remedy performance bottlenecks.
4. The adaptive computing resource allocation method for virtual network functions according to claim 2 , wherein the direct allocation method is to calculate a ratio of the computing resource requirement among each of the virtual network functions, and allocating total computing resources of a system to the virtual network functions according to the ratio.
5. The adaptive computing resource allocation method for virtual network functions according to claim 1 , wherein the adaptive computing resource allocation method uses a symmetric multi-processor (SMP) virtual machine (VM) and a multi-queue network card.
6. The adaptive computing resource allocation method for virtual network functions according to claim 1 , wherein an underlying logic of a virtual machine monitor (VMM) scheduler is not modified, and an unbalanced computing resource allocation is realized by means of a task priority mechanism provided by the VMM scheduler.
7. The adaptive computing resource allocation method for virtual network functions according to claim 6 , wherein by adjusting a priority of an underlying virtual central processing unit (vCPU) thread, a task obtains more or less computing resources, implementing an allocation of different amounts of the computing resources to an upper-level virtual network function.