Cluster load balance method transparent for operating system

Abstract

The present invention provides a cluster load balancing method transparent to an operation system. Main functional modules comprise a load balancing module, a processor migrating module and a communication module. The method is characterized by comprising the following steps: 1, driving a virtual processor to migrate; 2, driving balance migration; 3, sending a migrating request to a target node and negotiating; 4, storing and restoring a state of the virtual processor; and 5, communicating. The method better solves the problem of low resource utilization rate of a cluster system. Along with the development of more popularization of the cluster system and the continuous development of hardware virtualization technology in the future, the method can be a good solution for the low resource utilization rate of the cluster system and has good application prospect.

Description

(1) Technical field [0001] The present invention mainly relates to a cluster load balancing method that is transparent to the operating system, specifically refers to computer hardware virtualization technology and single system image technology, through virtualization technology, computing tasks are reasonably allocated to different processors, thereby improving the cluster system. resource utilization. belongs to the computer field. (2) Background technology [0002] 1. The load balancing technology of the current cluster system [0003] A cluster system is a single computer system formed by interconnecting a group of independent computers through a high-speed communication network. Its advantages lie in its system resource availability, scalability and parallelism, so it has been widely used in the field of parallel processing. [0004] As a server structure with good scalability and good performance, the cluster system has good cost performance. However, due to the lo...

AI Technical Summary

Problems solved by technology

At present, there are few implementation schemes at the system software layer, such as MOSIX, which implements load balancing at this layer and is more transparent to users, but it is difficult to implement, and some implementation schemes will bring other problems, such as MOSIX improves the resource utilization of the system to a certain extent through the mechanism of dynamic process migration, but the control structure after the process migration still remains on the host node, and the I / O operation is still completed on the host node, which brings a lot of network traffic. communication, reducing system performance

There are many schemes for implementing load balancing strategies at the middleware layer, and the implementation methods are different. They all support the requirements of distributed application load balancing from different aspects. The methods adopted depend on the specific application environment and are more versatile. Difference

For example, stateless distributed applications that need to balance workloads often use a load balancing service mechanism integrated with name services. This load balancing mechanism only supports static non-adaptive load balancing and cannot meet the needs of complex load balancing applications.

In addition, as the topology of distributed applications becomes increasingly complex and the scale of applications increases, the load balancing function of the existing middleware layer can no longer fully meet the scalability requirements of complex distributed applications. In terms of scalability, the middleware layer The load balancing mechanism does not have the advantage of

At the application layer, the most typical solution is to use Linux Virtual Servers (Linux Virtual Servers, LVS) as a load balancing solution for the cluster system. The load balancing at this layer is too closely combined with specific applications. Therefore, similar solutions are not universal. sex

Images