Applies To:

Show Versions Show Versions

Manual Chapter: Load Balancing with the Global Traffic Manager
Manual Chapter
Table of Contents   |   << Previous Chapter   |   Next Chapter >>

When the Global Traffic Manager receives a name resolution request, the system employs a load balancing mode to determine the best available virtual server to which to send the request. Once the Global Traffic Manager identifies the virtual server, it constructs a DNS answer and sends that answer back to the requesting clients local DNS server. The DNS answer, or resource record, can be either an A record that contains the IP address of the virtual server, or a CNAME record that contains the canonical name for a DNS zone.
Within the Global Traffic Manager, there are two categories of load balancing modes from which to select: static and dynamic. A static load balancing mode selects a virtual server based on a pre-defined pattern. A dynamic load balancing mode selects a virtual server based on current performance metrics.
The Global Traffic Manager provides a tiered load balancing system. A tiered load balancing system is a load balancing system that occurs at more than one point during the resolution process. The tiers within the Global Traffic Manager are as follows:
Wide IP-level load balancing
A wide IP contains two or more pools. The Global Traffic Manager load balances requests, first to a specific pool, and then to a specific virtual server in the selected pool. If the preferred, alternate, and fallback load balancing methods that are configured for the pool or virtual server fail, then the requests fail, or the system falls back to DNS.
Pool-level load balancing
A pool contains one or more virtual servers. After the Global Traffic Manager uses wide IP-level load balancing to select the best available pool, it uses a pool-level load balancing to select a virtual server within that pool. If the first virtual server within the pool is unavailable, the Global Traffic Manager selects the next best virtual server based on the load balancing mode assigned to that pool.
For each pool that you manage, the Global Traffic Manager supports three types of load balancing methods: preferred, alternate, and fallback. The preferred load balancing method is the load balancing mode that the system attempts to use first. If the preferred method fails to provide a valid resource, the system uses the alternate load balancing method. Should the alternate load balancing method also fail to provide a valid resource, the system uses the fallback method.
One of the key differences between the alternate methods and the other two load balancing methods is that only static load balancing modes are available from the alternate load balancing list. This limitation exists because dynamic load balancing modes, by definition, rely on metrics collected from different resources. If the preferred load balancing mode does not return a valid resource, it is likely that the Global Traffic Manager was unable to acquire the proper metrics to perform the load balancing operation. By limiting the alternate load balancing options to static methods only, the Global Traffic Manager can better ensure that, should the preferred method prove unsuccessful, the alternate method returns a valid result.
Table 7.1 shows a list of the supported static load balancing modes. Table 7.2 shows a list of the supported dynamic load balancing modes. Both tables indicate where you can use each mode in the Global Traffic Manager configuration. The following sections in this chapter describe how each load balancing mode works.
Use for wide IP load balancing
Use for wide IP load balancing
Static load balancing modes distribute connections across the network according to predefined patterns, and take server availability into account. The Global Traffic Manager supports the following static load balancing modes:
The None and Return to DNS load balancing modes are special modes that you can use to skip load balancing under certain conditions. The other static load balancing modes perform true load balancing as described in the following sections.
When you specify the Drop Packet load balancing mode, the Global Traffic Manager does nothing with the packet, and simply drops the request.
F5 Networks recommends that you use the Drop Packet load balancing mode only for the fallback method. The Global Traffic Manager uses the fallback method when the preferred and alternate load balancing modes do not provide at least one virtual server to return as an answer to a query.
When you specify the Fallback IP load balancing mode, the Global Traffic Manager returns the IP address that you specify as the fallback IP, as an answer to the query. Note that you can specify both an IPv4 and an IPv6 address as the fallback IP address. The IP address that you specify is not monitored for availability before being returned as an answer. When you use the Fallback IP mode, you can specify a disaster recovery site to return when no load balancing mode returns an available virtual server. F5 Networks recommends that you use the Fallback IP load balancing mode only for the fallback method. The Global Traffic Manager uses the fallback method when the preferred and alternate load balancing modes do not provide at least one virtual server to return as an answer to a query.
The Global Availability load balancing mode uses the virtual servers included in the pool in the order in which they are listed. For each connection request, this mode starts at the top of the list and sends the connection to the first available virtual server in the list. Only when the current virtual server is full or otherwise unavailable does Global Availability mode move to the next virtual server in the list. Over time, the first virtual server in the list receives the most connections and the last virtual server in the list receives the least number of connections.
The None load balancing mode is a special mode you can use if you want to skip the current load balancing method, or skip to the next pool in a multiple pool configuration. For example, if you set an alternate method to None in a pool, the Global Traffic Manager skips the alternate method and immediately tries the load balancing mode specified as the fallback method. If the fallback method is set to None, and you have multiple pools configured, the Global Traffic Manager uses the next available pool.
Tip: If you do not want the Global Traffic Manager to return multiple addresses that are potentially unavailable, F5 Networks recommends that you set the alternate method to Drop Packet.
You can also use the mode to limit each pool to a single load balancing mode. For example, you can set the preferred method in each pool to the desired load balancing mode, and then you can set both the alternate and fallback methods to None in each pool. If the preferred method fails, the None mode in both the alternate and fallback methods forces the Global Traffic Manager to go to the next pool for a load balancing answer.
The Ratio load balancing mode distributes connections among a pool of virtual servers as a weighted round robin. Weighted round robin refers to a load balancing pattern in which the Global Traffic Manager rotates connection requests among several resources based on a priority level, or weight, assigned to each resource. For example, you can configure the Ratio mode to send twice as many connections to a fast, new server, and only half as many connections to an older, slower server.
The Ratio load balancing mode requires that you define a ratio weight for each virtual server in a pool, or for each pool if you are load balancing requests among multiple pools. The default ratio weight for a server or a pool is set to 1.
The Return to DNS load balancing mode is another special mode that you can use to immediately return connection requests to the Local DNS for resolution. This mode is particularly useful if you want to temporarily remove a pool from service, or if you want to limit a pool in a single pool configuration to only one or two load balancing attempts.
The Round Robin load balancing mode distributes connections in a circular and sequential pattern among the virtual servers in a pool. Over time, each virtual server receives an equal number of connections.
The Static Persist load balancing mode uses the persist mask with the LDNS source IP address in a deterministic algorithm to map to a specific pool member in a pool. Unless the order or number of the pool members changes, the LDNS address is always mapped to the same pool member. Note that the pool and wide IP configurations must be the same on all Global Traffic Manager systems.
The Topology load balancing mode allows you to direct or restrict traffic flow by adding topology records to a topology statement in the configuration file. When you use the Topology load balancing mode, you can develop proximity-based load balancing. For example, a client request in a particular geographic region can be directed to a data center or server within that same region. The Global Traffic Manager determines the proximity of servers by comparing location information derived from the DNS message to the topology records.
This load balancing mode requires you to do some advanced configuration planning, such as gathering the information you need to define the topology records. The Global Traffic Manager contains an IP classifier that accurately maps local DNS servers, so when you create topology records, you can refer to continents and countries, instead of IP subnets.
See Chapter 9, Load Balancing Connection Requests Using Topologies, for detailed information about working with this and other topology features.
Dynamic load balancing modes distribute connections to servers that show the best current performance. The performance metrics taken into account depend on the particular dynamic mode you are using.
All dynamic load balancing modes make load balancing decisions based on the metrics collected by the big3d agents running in each data center. The big3d agents collect the information at set intervals that you define when you set the global timer variables. If you want to use the dynamic load balancing modes, you must run one or more big3d agents in each of your data centers, to collect the required metrics.
The Completion Rate load balancing mode selects the virtual server that currently maintains the least number of dropped or timed-out packets during a transaction between a data center and the client LDNS.
The CPU load balancing mode selects the virtual server that currently has the most CPU processing time available to handle name resolution requests.
The Hops load balancing mode is based on the traceroute utility, and tracks the number of intermediate system transitions (router hops) between a client LDNS and each data center. Hops mode selects a virtual server in the data center that has the fewest router hops from the Local DNS.
The Kilobytes/Second load balancing mode selects a virtual server that is currently processing the fewest number of kilobytes per second. You can use this mode only with servers for which the Global Traffic Manager can collect the kilobytes per second metric. See Chapter 13, Collecting Metrics, for details on the metrics the Global Traffic Manager collects.
The Least Connections load balancing mode is used for load balancing to virtual servers managed by a load balancing server, such as a Local Traffic Manager. The Least Connections mode selects the virtual server that currently hosts the fewest connections.
The Packet Rate load balancing mode selects a virtual server that is currently processing the fewest number of packets per second.
The Quality of Service load balancing mode uses current performance information to calculate an overall score for each virtual server, and then distributes connections based on each virtual servers score. The performance factors that the Global Traffic Manager takes into account include:
Figure 7.1 illustrates the equation that Global Traffic Manager uses to calculate the overall QoS score.
QoS_RTT * (QoSFACTOR_RTT / rtt) * 10 +
QoS_HOPS * (QoSFACTOR_HOPS / hops) +
QoS_HIT_RATIO * (hit_ratio / QoSFACTOR_HIT_RATIO) +
QoS_PACKET_RATE * (QoSFACTOR_PACKET_RATE / packet_rate) * 100 +
QoS_BPS * (QoSFACTOR_BPS / bps ) +
QoS_TOPOLOGY * (topology/ QoSFACTOR_TOPOLOGY) +
QoS_VS_CAPACITY * vs_capacity_score +
QoS_VS_SCORE * vs_score + 10 *
QoS_LCS * lcs_score
The Quality of Service load balancing mode is a customizable load balancing mode. For simple configurations, you can easily use this load balancing mode with its default settings. For more advanced configurations, you can specify different weights for each performance factor in the equation.
You can also configure the Quality of Service load balancing mode to use the dynamic ratio feature. With the dynamic ratio feature turned on, the Quality of Service mode becomes similar to the Ratio mode, where the connections are distributed in proportion to ratio weights assigned to each virtual server. The ratio weights are based on the QoS scores: the better the score, the higher percentage of connections the virtual server receives.
The Round Trip Times (RTT) load balancing mode selects the virtual server with the fastest measured round trip time between a data center and a client LDNS.
The Virtual Server Score load balancing mode instructs the Global Traffic Manager to assign connection requests to virtual servers based on a user-defined ranking system. This load balancing mode is available only for managing connections between virtual servers controlled by Local Traffic Manager systems.
Unlike other settings that affect load balancing operations, you cannot assign a virtual server score to a virtual server through the Global Traffic Manager. Instead, you assign this setting through the Local Traffic Manager that is responsible for the virtual server. For more information, see the F5 DevCentral web site: http://devcentral.f5.com.
The VS Capacity load balancing mode creates a list of the virtual servers, weighted by capacity, then picks one of the virtual servers from the list. The virtual servers with the greatest capacity are picked most often, but over time all virtual servers are returned. If more than one virtual server has the same capacity, then the Global Traffic Manager load balances using the Round Robin mode among those virtual servers.
The Quality of Service mode is a dynamic load balancing mode that includes a configurable combination of the Round Trip Time (RTT), Completion Rate, Packet Rate, Topology, Hops, Link Capacity, VS Capacity, and Kilobytes/Second (KBPS) modes. The Quality of Service mode is based on an equation that takes each of these performance factors into account.
When the Global Traffic Manager selects a virtual server, it chooses the server with the best overall score. In the event that one or more resources has an identical score based on the Quality of Service criteria, the Global Traffic Manager load balances connections between those resources using the Round Robin methodology. If the system cannot determine a Quality of Service score, it load balances connections across all pool members using the Round Robin load balancing mode, as well.
The Quality of Service mode has default settings that make it easy to use. There is no need to customize this mode; however, you can change the equation to put more or less weight on each individual factor. The following topics explain how to use and adjust the various settings of this mode.
Table 7.3 lists each Quality of Service (QoS) coefficient, its scale, a likely upper limit for each, and whether a higher or lower value is more efficient.
Example
upper limit
Percentage of successfully transferred packets (0-100%)
Score that defines network proximity by comparing server and LDNS IP addresses (0-232)
Scale
The raw metrics for each coefficient are not on the same scale. For example, completion rate is measured in percentages, while the packet rate is measured in packets per second.
Normalization
The Global Traffic Manager normalizes the raw metrics to values in the range of 0 to10. As the QoS value is calculated, a high measurement for completion rate is good, because a high percentage of completed connections are being made, but a high value for packet rate is not desirable because the packet rate load balancing mode attempts to find a virtual server that is not overly taxed at the moment.
Emphasis
You can adjust coefficients to emphasize one normalized metric over another. For example, consider the following QoS configuration:
In this configuration, if the completion rates for two virtual servers are close, the system chooses the virtual server with the best packet rate. If both the completion rates and the packet rates are close, the round trip time (RTT) breaks the tie. In this example, the metrics for Topology, Hops, Link Capacity, VS Capacity, and Kilobytes/Second modes are not used in determining how to distribute connections.
Note: You cannot set a value for both the Round Trip Time and Hops settings simultaneously. In situations where the Global Traffic Manager has a value for both settings, the round trip time is incorporated, while the value for the Hops setting is reset to 0.
You can establish your own custom settings for the Quality of Service load balancing method; however, you can customize the Quality of Service equation only at the pool level.
1.
On the Main tab of the navigation pane, expand Global Traffic and click Pools.
The main screen for pools opens.
2.
Click the name of the pool for which you want to modify the QoS equation.
The properties screen for that pool opens.
3.
On the menu bar, click Members.
The members screen opens.
4.
From either the Preferred or Fallback list, select Quality of Service.
6.
Click the Update button to save your changes.
The dynamic load balancing modes also support the Dynamic Ratio option. When you activate this option, the Global Traffic Manager treats dynamic load balancing values as ratios, and it uses each server in proportion to the ratio determined by this option. When the Dynamic Ratio option is disabled, the Global Traffic Manager uses only the server with the best result based on the dynamic load balancing mode you implemented (in which case it is a winner-takes-all situation), until the metrics information is refreshed.
Note: By default, the Dynamic Ratio setting is disabled (cleared).
To illustrate how the Dynamic Ratio setting works, consider a pool, primaryOne, that contains several pool members. This pool is configured so that the Global Traffic Manager load balances name resolution requests based on the Round Trip Time load balancing mode. The primaryOne pool contains two pool members: memberOne and memberTwo. For this example, the Global Traffic Manager determines that the round trip time for memberOne is 50 microseconds, while the round trip time for memberTwo is 100 microseconds.
If the primaryOne pool has the Dynamic Ratio setting disabled (the default setting), the Global Traffic Manager always load balances to the pool with the best value. In this case, this results in requests going to memberOne, because it has the lowest round trip time value.
If the primaryOne pool has the Dynamic Ratio setting enabled, however, the Global Traffic Manager treats the round trip time values as ratios and divide requests among pool members based on these ratios. In this case, this results in memberOne getting twice as many connections as memberTwo, because the round trip time for memberOne is twice as fast as the round trip time for memberTwo. Note that, with the Dynamic Ratio option enabled, both pool members are employed to handle connections, while if the option is disabled, only one pool member receives connections.
1.
On the Main tab of the navigation pane, expand Global Traffic and click Pools.
The main screen for pools opens.
2.
Click the name of the pool for which you want to enable the Dynamic Ratio option.
The properties screen for the pool opens.
3.
From the Configuration list, select Advanced.
Additional fields display with default settings.
4.
Check the Dynamic Ratio box.
5.
Click the Update button to save your changes.
Wide IP
When you define a wide IP, and you have multiple pools in your wide IP, you specify which load balancing mode to use in selecting a pool in the wide IP. To configure load balancing for a wide IP, see Configuring load balancing methods for wide IPs, following.
Pool
After the Global Traffic Manager selects a pool of virtual servers, it then employs the settings you specified as the preferred, alternate, and fallback load balancing methods to select a virtual server within the selected pool. To configure load balancing for a pool, see Configuring load balancing methods for pools.
There may be situations (for example, e-commerce, and other sites with multiple services) where you need to configure a wide IP so that connections are not sent to a given address unless multiple ports or services are available. You configure this behavior after you define the wide IP. For details, see Employing additional load balancing options.
The Global Traffic Manager supports a wide variety of load balancing methods for distributing network connection requests across the pools in a wide IP. For information on these load balancing methods, see Understanding load balancing on the Global Traffic Manager.
1.
On the Main tab of the navigation pane, expand Global Traffic and click Wide IPs.
The main screen for wide IPs opens.
2.
Click the name of the wide IP for which you want to configure a load balancing method.
The properties screen for the wide IP opens.
3.
On the menu bar, click Pools.
The pools screen opens displaying a list of the pools currently assigned to the wide IP.
4.
Select the appropriate load balancing options.
For additional information on these load balancing options, see the online help.
5.
Click the Update button to save your changes to the wide IP.
The Global Traffic Manager supports a wide variety of load balancing methods for distributing network connection requests across the virtual servers in a pool. For information on these load balancing modes, see Understanding load balancing on the Global Traffic Manager.
For each pool that you manage, the Global Traffic Manager supports three types of load balancing methods: preferred, alternate, and fallback. The preferred load balancing method is the load balancing method that the system attempts to use first. If the preferred method fails to provide a valid resource, the system uses the alternate load balancing method. Should the alternate load balancing method also fail to provide a valid resource, the system uses the fallback method.
1.
On the Main tab of the navigation pane, expand Global Traffic and click Pools.
The main screen for pools opens.
2.
Click the name of the pool for which you want to configure load balancing methods.
The properties screen for that pool opens.
3.
On the menu bar, click Members.
The members screen opens displaying a list of the virtual servers currently assigned to the pool.
4.
Select the appropriate load balancing options.
For additional information on these load balancing options, see the online help.
5.
Click the Update button to save your changes to the pool.
The Global Traffic Manager supports three types of load balancing methods at the pool level: preferred, alternate, and fallback. The preferred load balancing method is the load balancing method that the system attempts to use first. If the preferred method fails to provide a valid resource, the system uses the alternate load balancing method. Should the alternate load balancing method also fail to provide a valid resource, the system uses the fallback method.
The fallback load balancing method is unique among the three load balancing method that you can apply to a pool. Unlike the preferred and alternate methods, the fallback method ignores the availability status of a resource. This occurs to ensure that the Global Traffic Manager returns a response to the DNS request. For more information on the determining resource health and availability, see Chapter 8, Managing Connections.
Note: If you do not want the Global Traffic Manager to return an address that is potentially unavailable, F5 Networks recommends that you set the alternate load balancing method to Drop Packet.
The Global Traffic Manager contains several options that help you control how the system responds when using a fallback load balancing setting. These options allow you to:
When you assign a load balancing mode to the fallback load balancing method for a pool, the Global Traffic Manager uses the mode differently than for the preferred and alternate methods. With the fallback load balancing method, the Global Traffic Manager load balances the name resolution request after verifying that the virtual server address returned is up or down. However, unlike with other load balancing methods, you can opt to use the fallback load balancing method to resolve a name resolution request without verifying the status of the virtual server.
Note: By default, the Respect Fallback Dependency option is disabled. When you enable it, the system verifies that the virtual server is available for using it for fallback load balancing.
1.
On the Main tab of the navigation pane, expand System and click Configuration.
The general properties screen opens.
2.
From the Global Traffic menu, choose Load Balancing.
The load balancing properties screen opens.
3.
Verify that the Respect Fallback Dependency box is clear.
4.
Click the Update button to save your changes.
The Global Traffic Manager supports additional options that affect how the system load balances name resolution requests. These options are:
The Ignore Path TTL option instructs the Global Traffic Manager to use path information gathered during metrics collection even if the time-to-live value for that information has expired. This option is often used when you want the system to continue using a dynamic load balancing mode even if some metrics data is temporarily unavailable, and you want the Global Traffic Manager to use old metric data rather than employ an alternate load balancing method. This option is disabled by default.
The Verify Virtual Server Availability option instructs the Global Traffic Manager to verify that a virtual server is available before returning it as a response to a name solution request. If this option is disabled, the system responds to a name resolution request with the virtual servers IP address regardless of whether the server is up or down. This option is rarely deactivated outside of a test or staging environment, and is enabled by default.
1.
On the Main tab of the navigation pane, expand System and click Configuration.
The general properties screen opens.
2.
From the Global Traffic menu, choose Load Balancing.
The load balancing properties screen opens.
3.
Enable or disable the Ignore Path TTL and Verify Virtual Server Availability options as needed.
4.
Click the Update button to save your changes.
Table of Contents   |   << Previous Chapter   |   Next Chapter >>

Was this resource helpful in solving your issue?




NOTE: Please do not provide personal information.



Incorrect answer. Please try again: Please enter the words to the right: Please enter the numbers you hear:

Additional Comments (optional)