WO2015123985A1

WO2015123985A1 - Method and device for adjusting load balancing, and computer storage medium

Info

Publication number: WO2015123985A1
Application number: PCT/CN2014/085865
Authority: WO
Inventors: 徐方明
Original assignee: 中兴通讯股份有限公司
Priority date: 2014-02-24
Filing date: 2014-09-03
Publication date: 2015-08-27
Also published as: CN104869076A

Abstract

Disclosed are a method and device for adjusting load balancing. The method comprises: according to link conditions, configuring priorities for various links; and according to the priorities of all links, forwarding the total traffic of current links in sequence in a high to low order of the priorities, to adjust load balancing. By means of the present invention, priorities are configured for various links, priorities of links can be re-configured according to the change of link conditions, and the total traffic of current links is forwarded according to the sequence of the priorities, so that the load balancing can be dynamically conducted, thereby improving the link transmission quality, and solving the problem of poor load balancing of traffic due to the fact that the static load balancing cannot be adjusted in accordance with bandwidth demands of various links and configuration or link traffic, etc. Also disclosed at the same time is a computer storage medium.

Description

Method and device for adjusting load balancing and computer storage medium

Technical field

The present invention relates to the field of communications, and in particular, to a method and apparatus for adjusting load balancing and a computer storage medium.

Background technique

Multiple data links between two data communication devices in an IP network are generally deployed to form load balancing. The sum of bandwidths between links is the traffic value that can be assumed between devices. The bandwidth carried by each link is fixed. When the total traffic of the link increases suddenly, the total number of outgoing traffic of the link between the links is greater than that of the link load balancing. The phenomenon of the package. When the egress link of the user network adopts dual-link uplink, and one of the links is a leased transmission link, the load between the two links can be calculated based on the interface bandwidth to calculate the load weight of the load balancing. Equilibrium, and hope to make full use of the leased transmission link, dynamically adjust the weight of the load balancing according to the traffic on the leased link, and perform the sharing of each stream. Since the actual bandwidth of the leased transmission link is generally smaller than the physical bandwidth of the interface, the proportional weight of the load balancing needs to be calculated according to its actual bandwidth. The egress link of the user network adopts dual-link uplink. One of them serves as the primary link and the other serves as the backup. The backup link may be used for other service requirements and occupy part of the backup link bandwidth. When the active link traffic exceeds the bandwidth, the standby link needs to participate in the active link without affecting other services to implement load balancing.

However, in the prior art, static load balancing is used, and a fixed mode is used for load balancing. However, static load balancing cannot solve the problem that the bandwidth of each link needs to be adjusted according to configuration or link traffic, resulting in traffic load. The balance is poor. However, there is no method in the prior art that can adjust according to the configuration and link traffic according to the bandwidth requirement of each link.

Summary of the invention

In view of the above, the embodiments of the present invention provide a method and a device for adjusting load balancing. It is expected that static load balancing cannot be adjusted according to configuration or link traffic according to bandwidth requirements of each link, resulting in load balancing of traffic. Poor question.

In order to achieve the above objective, the first aspect of the present invention provides a method for adjusting load balancing, including: configuring a priority of each link according to a link condition; and according to a priority of all links, according to a priority from high to low The order in turn forwards the total traffic of the current link to adjust the load balancing.

Preferably, before forwarding the total traffic of the current link in descending order of priority according to the priority of all the links, the method further includes: counting the traffic of each link to determine the total traffic of the current link.

Preferably, forwarding the total traffic of the current link in order of priority from highest to lowest includes: detecting whether the bandwidth of the link with the highest priority can forward the total traffic of the current link; and the highest priority If the bandwidth of the link cannot be completely forwarded, the remaining traffic is forwarded on the link corresponding to the next priority of the link with the highest priority; if the next priority corresponds to the link If the remaining bandwidth is still not fully forwarded, the remaining traffic after forwarding is forwarded in order of priority until the total traffic is forwarded.

Preferably, the method further includes: when there are multiple links with the same priority, the traffic to be forwarded is allocated and forwarded according to the bandwidth ratio of the link.

Preferably, after the total traffic of the current link is sequentially forwarded in descending order of priority, the method further includes: detecting a total traffic of the current link according to a predetermined time interval; and in case the total traffic is changed, according to each The forwarding capability of the link recalculates the bandwidth of each link, wherein the forwarding capability is: the product of the physical bandwidth and the waterline, or the product of the physical bandwidth and the waterline and the load balancing bandwidth.

The second aspect of the present invention further provides an apparatus for adjusting load balancing, including: a configuration module configured to configure a priority of each link according to a link condition; and a forwarding module configured to According to the priority of all links, the total traffic of the current link is forwarded in order of priority from high to low to adjust load balancing.

Preferably, the apparatus further includes: a statistics module configured to count traffic of each link to determine a total traffic of the current link.

Preferably, the forwarding module includes: a detecting unit configured to detect whether a bandwidth of a link with the highest priority can forward the total traffic of the current link; and a forwarding unit configured to be in the link with the highest priority If the bandwidth cannot be completely forwarded, the remaining traffic is forwarded on the link corresponding to the next priority of the link with the highest priority; the bandwidth of the link corresponding to the next priority is still When the remaining traffic cannot be completely forwarded, the traffic remaining again after the forwarding is sequentially forwarded in the priority order until the total traffic is completely forwarded.

Preferably, the forwarding module is further configured to allocate and forward the traffic to be forwarded according to the bandwidth ratio of the link, if there are multiple links with the same priority.

Preferably, the apparatus further includes: a detecting module configured to detect a total traffic of the current link according to a predetermined time interval; and a determining module configured to, according to the forwarding capability of each link, restart in the case that the total traffic changes The bandwidth of each link is calculated, wherein the forwarding capability is: the product of the physical bandwidth and the watermark, or the product of the physical bandwidth and the waterline and the load balancing bandwidth.

The third aspect of the present invention further provides a computer storage medium, where the computer storage medium stores computer executable instructions for performing at least one of the methods of the first aspect.

In the embodiment of the present invention, the priority of each link is configured, and the priority of the link can be re-configured according to the change of the link status, and the total traffic of the current link is forwarded according to the priority order, so that load balancing can be dynamically performed. The transmission quality of the link is improved, and the static load balancing cannot be adjusted according to the configuration of the link or the link traffic according to the bandwidth requirements of each link. The problem of poor load balancing for traffic.

DRAWINGS

1 is a flowchart of a method for adjusting load balancing according to an embodiment of the present invention;

2 is a schematic diagram of a first structure of an apparatus for adjusting load balancing according to an embodiment of the present invention;

3 is a schematic diagram of a second structure of an apparatus for adjusting load balancing according to an embodiment of the present invention;

4 is a schematic structural diagram of a device forwarding module for adjusting load balancing according to an embodiment of the present invention;

FIG. 5 is a third schematic structural diagram of an apparatus for adjusting load balancing according to an embodiment of the present invention; FIG.

6 is a flowchart of dynamically adjusting load balancing according to an embodiment of the present invention;

FIG. 7 is a fourth schematic structural diagram of an apparatus for adjusting load balancing according to an embodiment of the present invention.

detailed description

The preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings.

An embodiment of the present invention provides a method for adjusting load balancing. The process of the method is as shown in FIG. 1 and includes steps S102 to S104:

S102. Configure a priority of each link according to a link condition.

The link status may include various factors, such as whether the current link is a leased link, the bandwidth of the current link, the bandwidth occupied by the current link, and the like. When setting, if the current link is set for the leased link, the priority of the leased link can be set to high, the priority of other links is set to low, and further, it can be further set to low. Prioritize the priority of the other link status, for example, configure the bandwidth. Larger links have a higher priority.

S104. Forward the total traffic of the current link according to the priority of all links according to the priority from high to low to adjust load balancing.

In the embodiment of the present invention, the priority of each link is configured, and the priority of the link can be re-configured according to the change of the link status, and the total traffic of the current link is forwarded according to the priority order, so that load balancing can be dynamically performed. The transmission quality of the link is improved, and the problem that the static load balancing cannot be adjusted according to the configuration or link traffic of the bandwidth of each link may be solved, resulting in a problem of poor load balance of the traffic.

During the implementation process, if a link is damaged or the service is changed, the priority and watermark can be adjusted according to the actual situation. For example, if the link with priority 7 is damaged, you can adjust the priority configuration and set its priority to 1 or 0. In this way, you can forward according to the reconfigured priority. Otherwise, if the current link If the service is changed from the previous multi-service to the single service, the watermark corresponding to the link can be changed for forwarding.

Before forwarding the total traffic of the current link in the order of priority from high to low according to the priority of all links, the traffic of each link needs to be counted to determine the total traffic of the current link.

During the implementation process, the total traffic of the current link can be forwarded in the order of priority from high to low. The following process can be performed: First, check whether the bandwidth of the link with the highest priority can forward the total traffic of the current link; If the bandwidth of the link with the highest priority cannot be used to forward the total traffic, the remaining traffic is forwarded on the link corresponding to the next priority of the link with the highest priority. The remaining traffic can be forwarded. If the bandwidth of the link corresponding to the next priority cannot be completely forwarded, the traffic remaining after forwarding will be forwarded in the order of priority until the total traffic is forwarded.

If there are multiple links with the same priority, the traffic to be forwarded can be allocated and forwarded according to the bandwidth ratio of the link. For example, the bandwidth ratio of the three links is 4:3:1. The traffic to be forwarded can be equalized according to this ratio, and then forwarded.

After forwarding the total traffic of the current link in the order of priority from high to low, the total traffic of the current link needs to be detected according to a predetermined time interval (ie, a sampling period), and the predetermined time interval may be set according to different time periods. For example, it can be set shorter when the sampling period is just sent. When the flow is stable, the sampling period can be set longer. In the case of a total traffic change, the bandwidth of each link is recalculated according to the forwarding capability of each link, where the forwarding capability is: the product of the physical bandwidth and the waterline, or the product of the physical bandwidth and the waterline and the load balancing bandwidth. The smallest one.

The embodiment of the present invention further provides an apparatus for adjusting load balancing. The structure of the apparatus is shown in FIG. 2, and includes: a configuration module 10 configured to configure a priority of each link according to a link condition; and a forwarding module 20, The configuration is coupled to the configuration module 10, and configured to forward the total traffic of the current link in order of priority according to priorities of all links to adjust load balancing.

FIG. 3 is a schematic structural diagram of the foregoing apparatus. The apparatus may further include: a statistics module 30 coupled to the forwarding module 20 and configured to collect traffic of each link to determine a total traffic of the current link. In the implementation, the forwarding module 20 may be configured to allocate and forward the traffic to be forwarded according to the bandwidth ratio of the link when there are multiple links with the same priority.

FIG. 4 is a schematic structural diagram of the foregoing forwarding module 20, including: a detecting unit 202 configured to detect whether a bandwidth of a link with the highest priority can forward the total traffic of the current link; and a forwarding unit 204 coupled with the detecting unit 202, If the bandwidth of the link with the highest priority cannot be used to forward the total traffic, the remaining traffic is forwarded on the link corresponding to the next priority of the link with the highest priority; the link corresponding to the next priority If the bandwidth cannot be completely forwarded, the remaining traffic after forwarding will be forwarded in order of priority until the total traffic is forwarded.

FIG. 5 is a schematic diagram of a preferred structure of the apparatus for adjusting load balancing. On the basis of FIG. 3, the method further includes: a detecting module 40 coupled to the statistic module 30 and configured to be scheduled according to a predetermined time. Detecting the total traffic of the current link; the determining module 50 is coupled to the forwarding module 20, and configured to recalculate the bandwidth of each link according to the forwarding capability of each link when the total traffic changes, wherein the forwarding capability is : The smallest of the product of the physical bandwidth and the watermark, or the product of the physical bandwidth and the waterline and the load balancing bandwidth.

The embodiment of the invention provides a method and a device for dynamically adjusting load balancing, which can dynamically adjust the bandwidth of each link of the load balancing according to the network user demand and the link traffic change. The method is applicable to IPV4, IPV6 networks, and MPLS networks.

The device of the embodiment of the invention can configure a load balancing policy, a load balancing priority, a load balancing bandwidth, and a load balancing watermark. When the policy, priority, bandwidth, and waterline change, the load balancing is dynamically calculated. The outgoing traffic that the link actually forwards can also be obtained from the sub-interface. When the total traffic of the load balancing link changes, the load balancing is dynamically calculated. The sampling period can be set shorter when the traffic is just sent, ensuring load balancing and recalculating in time. When the traffic is stable, the sampling period can be set longer to reduce the CPU usage. The load balancing can also be used to dynamically calculate the bandwidth of the outbound interface of each link, so that traffic is preferentially higher than the priority link. After the capacity is capped, the remaining traffic goes through the sub-optimal priority link. The time-sharing mechanism is used to manage the dynamic load balancing calculation process to prevent the device from having a large amount of load balancing and dynamic calculation so that the CPU usage is short-lived.

The following description is made in conjunction with the accompanying drawings and specific embodiments.

The user sets the parameters related to load balancing through the network management or OAM interface in the network deployment. The following describes each parameter in detail.

(1) Link load balancing priority. A higher priority value indicates a higher priority. When a network user leases a link of a higher-level user, the link should be set to a higher priority (the link is of higher quality). Dynamic load balancing calculations are enabled only if all links are configured with a priority.

(2) Link load balancing bandwidth. The load balancing bandwidth cannot be greater than the physical bandwidth of the link. This parameter needs to be configured when the network user needs traffic to be allocated according to the expected ratio (for example, 4:2:1).

(3) Link load balancing waterline. The load balancing waterline defines the percentage of physical traffic that the traffic of the link can reach. When a load balancing service bears other services on a link, the watermark can determine how much bandwidth is allocated to load balancing.

As shown in FIG. 6 , after the user sets the parameters, the bandwidth of each link of the load balancing is dynamically adjusted according to parameters such as the traffic of the link, and the steps S601 to S610 are included.

S601: Configure dynamic load balancing parameters. If the load balancing policy, priority, bandwidth, waterline, and link traffic change, go to the process.

S602. Determine whether the current configuration meets the condition of dynamic load balancing. Dynamic calculations are only performed when all links are configured with load balancing priorities. If yes, step S603 is performed, otherwise step S610 is performed.

S603. Acquire a forwarding capability of each link, that is, a smaller one of physical bandwidth * watermark and load balancing bandwidth.

S604. The traffic of each link is counted, and the total traffic of the load balancing is obtained.

S605, whether there is a link has not calculated bandwidth. If yes, step S606 is performed, otherwise step S610 is performed.

S606. Acquire a high priority level in the configuration, and obtain a total forwarding capability of the current priority link.

S607. Determine whether the highest priority link can forward the total traffic on the link. If yes, step S608 is performed, otherwise step S609 is performed.

At S608, the traffic completely goes from the highest priority link, and other priority links do not take traffic. At this time, the bandwidth of the priority link is the total traffic, and the bandwidth of the remaining priority links is zero, and the process proceeds to step S610.

S609: If the forwarding is not complete, the traffic is shared according to the maximum forwarding capability of the link, and the remaining traffic is obtained, and the remaining traffic is switched to the sub-optimal priority link. And the process returns to step S605 to continue the execution. At this time, the bandwidth of the highest priority link is its forwarding capability.

S610, the process ends.

The process of continuing the foregoing is as follows: determining whether the sub-optimal priority link can forward the remaining total traffic on the link. If possible, the remaining traffic is completely removed from the sub-optimal priority link, and the priority is lower than the traffic on its link. At this time, the bandwidth of the secondary priority link is the remaining traffic, the priority is lower than the bandwidth of its link is zero, and the process ends. If the forwarding is not complete, the traffic is shared according to the maximum forwarding capability of the link, and the remaining traffic flows to the priority link again. At this time, the traffic of the sub-preferred priority link is its forwarding capability.

Determine whether the re-priority priority link can forward the remaining total traffic on the link. If possible, the remaining traffic is completely removed from the good priority link again. At this time, the bandwidth of the priority link again is the remaining traffic, and the priority is lower than the bandwidth of its link is zero, and the process ends. If the forwarding is not complete, the traffic is shared according to the maximum forwarding capability of the link, and the remaining traffic flows to the next priority link. At this time, the traffic of the priority priority road again is its forwarding capability. Repeat the above process until the remaining traffic is forwarded from a link of a certain priority level.

When the priority of the multiple links is the same, the same priority link distributes the traffic of the priority according to the forwarding capability. When the total load of the load balance changes and the load balancing parameters change. From step S601, the bandwidth of each link of the load balancing is recalculated.

Load balancing The bandwidth of the link can be dynamically adjusted according to the link traffic according to the above method. This can solve the problem of packet loss caused by a sudden increase in link traffic, and can also satisfy the problem that the uplink dual-link secondary carrier preferentially uses the leased first-level carrier link and the proportional use of the standby link.

The scenario in which the above traffic suddenly increases is similar to the processing in FIG. 6 above, and includes the following processing:

The total traffic on the link changes, first triggering the recalculation of the ECMP link bandwidth. Subsequently, similar to the execution process of FIG. 6, it is judged whether the highest priority link can accommodate the traffic. If it can be accommodated, the highest priority link bandwidth is adjusted for the link traffic at this moment, and the process ends. If not, The highest priority link bandwidth remains the same, and traffic flows to the sub-optimal priority link.

Determine if the sub-optimal priority can accommodate the remaining traffic. If yes, adjust the sub-optimal priority link bandwidth to the remaining traffic, and the process ends. Otherwise, the bandwidth of the sub-optimal priority link is unchanged, and the traffic flow is again prioritized. Since the total bandwidth of the link is greater than the link traffic, the traffic that overflows from the previous priority can always be accommodated in a certain priority link. The bandwidth of this priority link is the traffic that overflows from the previous priority link. The process ends.

In the above process, by adjusting the bandwidth of some links, it is ensured that when the traffic of the ECMO link suddenly becomes large and the total bandwidth of the link does not exceed the total bandwidth of the link, the traffic does not drop.

The scenario in which one of the uplink links is leased is similar to the process in Figure 6 above, and includes the following processing:

First, the ECMP priority of the leased link is set to the highest priority of 7. Subsequently, similar to the execution process of FIG. 6, it is determined whether the highest priority link can accommodate the total link traffic. When it can accommodate, the traffic no longer takes the sub-optimal priority link. The highest priority link bandwidth is the total link traffic, and the other priority link bandwidths are 0, and the process ends. When it can't be accommodated, traffic flows to sub-optimal priorities.

Determine if the sub-optimal priority can accommodate the remaining traffic. If yes, adjust the sub-optimal priority link bandwidth to the remaining traffic, and the process ends. Otherwise, the bandwidth of the sub-optimal priority link is unchanged, and the traffic flow is again prioritized.

Determine whether the sub-optimal priority link can forward the remaining total traffic on the link. If possible, the remaining traffic is completely removed from the sub-optimal priority link, and the priority is lower than the traffic on its link. At this time, the bandwidth of the secondary priority link is the remaining traffic, the priority is lower than the bandwidth of its link is zero, and the process ends. If the forwarding is not complete, the traffic is shared according to the maximum forwarding capability of the link, and the remaining traffic flows to the priority link again. At this time, the traffic of the sub-preferred priority link is its forwarding capability. Repeat the above process until the remaining traffic is forwarded from a link of a certain priority level. If the flow of the link changes, traffic is distributed again starting from the highest priority link.

In the above process, the leased link priority is set to a high priority to ensure that the leased link is fully utilized, and the traffic distribution is dynamically adjusted according to changes in attributes such as traffic.

The scenario in which one of the uplink dual-links carries other services is similar to the processing in Figure 6 above, and includes the following processing:

First, set the priority of carrying other service links to a lower priority, and then set the watermarks carrying other service links so that the bandwidth of other services does not exceed the physical bandwidth of the link. Subsequently, similar to the execution process of FIG. 6, it is judged whether the highest priority link can accommodate the total link traffic. When it can accommodate, the traffic no longer takes the sub-optimal priority link. The highest priority link bandwidth is the total link traffic, and the other priority link bandwidths are 0, and the process ends. The sub-preferred priority link only shares traffic when it cannot accommodate it.

Determine if the sub-optimal priority link can accommodate the remaining traffic. The criterion is whether the traffic size exceeds the forwarding capability of the link. If the traffic exceeds this value, the link will stop sharing even if the link can take on more traffic. At this time, the bandwidth of the sub-preferred priority link is its forwarding capability. The overflowed traffic flow is again prioritized. If not exceeded, the bandwidth of the link is the traffic of the link, and the process ends. Repeat the above process until the remaining traffic is forwarded from a link of a certain priority level. If the flow of the link changes, traffic is distributed again starting from the highest priority link.

In the above process, by lowering the priority of other service links and setting the corresponding waterline, the link can be guaranteed as an alternative link, and the preferred link can not participate in ECMP after the traffic is not carried, and does not affect the chain. Participate in dynamic calculations in the case of other businesses.

As shown in FIG. 7, another embodiment of an apparatus for adjusting load balancing is provided below, the apparatus including a processor 302, a storage medium 304, and at least one external communication interface 301; the processor 302, the storage medium 304, and an external Communication interfaces 301 are each connected by a bus 303. The processor 302 can be a processing component such as a microprocessor, a central processing unit, a digital signal processor, or a programmable logic array.

Computer-executable instructions are stored on the storage medium 304; the processor 302 executes the The computer executable instructions stored in the storage medium 304 can at least implement the following scheme:

Configure the priority of each link according to the link status;

According to the priority of all links, the total traffic of the current link is forwarded in order of priority from high to low to adjust load balancing.

The embodiment of the present invention further describes a computer storage medium, where the computer storage medium stores a computer program, and the computer program is used to execute at least one of the technical solutions in the method for adjusting load balancing according to the embodiment of the present invention. First, specifically adjust the load balancing method as shown in FIG.

The computer storage medium may be a removable storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and the like, which can store program codes. It is a non-transient storage medium.

The above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention. Modifications made in accordance with the principles of the invention are understood to fall within the scope of the invention.

Claims

A method of adjusting load balancing, including:

Configure the priority of each link according to the link status;

According to the priority of all links, the total traffic of the current link is forwarded in order of priority from high to low to adjust load balancing.
The method of claim 1, wherein before the total traffic of the current link is forwarded in descending order of priority according to the priority of all the links, the method further includes:

The traffic of each link is counted to determine the total traffic of the current link.
The method of claim 1, wherein the total traffic of the current link is sequentially forwarded in descending order of priority, including:

Detecting whether the bandwidth of the link with the highest priority can forward the total traffic of the current link;

If the bandwidth of the link with the highest priority cannot be all the total traffic, the remaining traffic is forwarded on the link corresponding to the next priority of the link with the highest priority;

If the bandwidth of the link corresponding to the next priority cannot be completely forwarded, the remaining traffic after forwarding is sequentially forwarded in priority order until the total traffic is forwarded.
The method of claim 3 wherein the method further comprises:

When there are multiple links with the same priority, the traffic to be forwarded is allocated and forwarded according to the bandwidth ratio of the link.
The method according to any one of claims 1 to 4, wherein after the total traffic of the current link is sequentially forwarded in descending order of priority, the method further includes:

Detecting the total traffic of the current link according to a predetermined time interval;

In the case that the total traffic changes, each of the link forwarding capabilities is recalculated The bandwidth of the link, wherein the forwarding capability is: the product of the physical bandwidth and the watermark, or the smallest of the product of the physical bandwidth and the waterline and the load balancing bandwidth.
A device for adjusting load balancing, including:

The configuration module is configured to configure the priority of each link according to the link status;

The forwarding module is configured to forward the total traffic of the current link according to the priority of all links according to the priority from high to low to adjust load balancing.
The apparatus of claim 6 further comprising:

The statistics module is configured to collect traffic of each link to determine the total traffic of the current link.
The apparatus of claim 6, wherein the forwarding module comprises:

The detecting unit is configured to detect whether the bandwidth of the link with the highest priority can forward the total traffic of the current link;

The forwarding unit is configured to: when the bandwidth of the link with the highest priority cannot fully forward the total traffic, perform the remaining traffic on the link corresponding to the next priority of the link with the highest priority If the bandwidth of the link corresponding to the next priority is still unable to forward the remaining traffic, the traffic remaining after the forwarding is forwarded in the order of priority until the total traffic is forwarded.
The apparatus of claim 6 wherein

The forwarding module is further configured to allocate and forward the traffic to be forwarded according to the bandwidth ratio of the link when there are multiple links with the same priority.
The apparatus according to any one of claims 6 to 9, further comprising:

The detecting module is configured to detect the total traffic of the current link according to a predetermined time interval;

The determining module is configured to recalculate the bandwidth of each link according to the forwarding capability of each link, where the total traffic is changed, where the forwarding capability is: a product of physical bandwidth and waterline, or physical The smallest of the product of bandwidth and watermark and load balancing bandwidth.
A computer storage medium storing computer executable in the computer storage medium A line of instructions for performing at least one of the methods of claims 1 to 5.