Round-robin scheduling

The term run-out method or English round-robin called a scheduling method, that is, it assigns several competing processes limited resources. The round-robin method allows all processes sequentially for each a short period - a so-called time slot - access to the required resources; this is called also arbitration.

Round -robin is used for load distribution (load balancing ). The aim of load balancing is to take several identical resources as evenly as possible.

Realization

The processes are managed in a queue. The foremost process gets a time slot long access to the resources, then he joins at the end of the queue and processes all move one position in front. The next process is selected according to the FIFO principle. The process can also voluntarily return before the resource. Even if a process is completed before the end of his time slot, the resources are allocated immediately.

Scheduling for multitasking operating system

For operating systems with preemptive multitasking scheduler for the active processes builds an execution plan after the round-robin method. Then he determined after each time slot on a queue the process that comes next in the series. The dispatcher then divides this process one time slot long the processor.

Example load balancing domain server

When load balancing round robin is for example used in the domain name system, in which a name server query can provide multiple IP addresses. For load balancing on large sites or IRC networks, this happens on multiple physical servers.

See also load balancing via DNS

Example load balancing routing

Routing protocols such as Routing Information Protocol (RIP) set the round-robin load balancing method to a various lines ( routes). Routes with the same metric node and the same destination network will be charged according to the package delivery of the series. This is done either for each forwarded packet individually ( per packet ) or for each new target host (by destination).

Assessment

The criteria on which this review is based, see Scheduling section "Criteria ".

Round - Robin treats all processes equally, so that on the one hand, no process is being treated unfairly or even starving, but on the other hand also is not possible, processes with higher urgency preferred work off. The throughput of this scheduling method is still very high in general neither particularly low. The use of time slots of fixed length does round-robin inflexible.

• Operating system theory
• Algorithm