Abstract: Some embodiments provide a method of performing load balancing for a group of machines that are distributed across several physical sites. The method of some embodiments iteratively computes (1) first and second sets of load values respectively for first and second sets of machines that are respectively located at first and second physical sites, and (2) uses the computed first and second sets of load values to distribute received data messages that the group of machines needs to process, among the machines in the first and second physical sites. The iterative computations entail repeated calculations of first and second sets of weight values that are respectively used to combine first and second load metric values for the first and second sets of machines to repeatedly produce the first and second sets of load values for the first and second sets of machines.

Abstract: Some embodiments provide a method of performing load balancing for a group of machines that are distributed across several physical sites. The method of some embodiments iteratively computes (1) first and second sets of load values respectively for first and second sets of machines that are respectively located at first and second physical sites, and (2) uses the computed first and second sets of load values to distribute received data messages that the group of machines needs to process, among the machines in the first and second physical sites. The iterative computations entail repeated calculations of first and second sets of weight values that are respectively used to combine first and second load metric values for the first and second sets of machines to repeatedly produce the first and second sets of load values for the first and second sets of machines.

Abstract: Example methods and systems for service-aware global server load balancing are described. One example may involve a first load balancer receiving, from a client device, a request to access a service associated with an application deployed in at least a first cluster and a second cluster. In response to determination that a first pool in the first cluster is associated with an unhealthy status, the first load balancer may identify a second pool implementing the service in the second cluster, the second pool being associated with a healthy status and includes one or more second backend servers selectable by a second load balancer to process the request. Failure handling may be performed by interacting with the client device, or the second load balancer, to allow the client device to access the service implemented by the second pool in the second cluster.

Abstract: Some embodiments provide a method of performing load balancing for a group of machines that are distributed across several physical sites. The method of some embodiments iteratively computes (1) first and second sets of load values respectively for first and second sets of machines that are respectively located at first and second physical sites, and (2) uses the computed first and second sets of load values to distribute received data messages that the group of machines needs to process, among the machines in the first and second physical sites. The iterative computations entail repeated calculations of first and second sets of weight values that are respectively used to combine first and second load metric values for the first and second sets of machines to repeatedly produce the first and second sets of load values for the first and second sets of machines.

Abstract: The method of synchronizes network address translation (NAT) records between an active gateway and a standby gateway. The method of some embodiments synchronizes NAT records of long-term data flows more frequently than those of short-term flows. Multiple data flows pass between a device at an internal source address and a device at an external destination address through the active NAT gateway. For each flow, the method generates a NAT record. The method then determines whether the data flow is a short-term flow or a long-term flow and synchronizes the NAT records of the long-term flows, but not the NAT records of the short-term flows, with the standby gateway. The method of some embodiments synchronizing NAT records more frequently when NAT records are being generated quickly relative to prior generation rates and less frequently when NAT records are being generated slowly relative to the prior generation rates.

Abstract: The method of synchronizes network address translation (NAT) records between an active gateway and a standby gateway. The method of some embodiments synchronizes NAT records of long-term data flows more frequently than those of short-term flows. Multiple data flows pass between a device at an internal source address and a device at an external destination address through the active NAT gateway. For each flow, the method generates a NAT record. The method then determines whether the data flow is a short-term flow or a long-term flow and synchronizes the NAT records of the long-term flows, but not the NAT records of the short-term flows, with the standby gateway. The method of some embodiments synchronizing NAT records more frequently when NAT records are being generated quickly relative to prior generation rates and less frequently when NAT records are being generated slowly relative to the prior generation rates.

Abstract: The method of some embodiments synchronizes NAT records between an active gateway and a standby gateway. The active gateway encodes a NAT record that includes at least an external source IP address. The encoded NAT record does not include the external source IP address but does include an identifier that uniquely specifies the external source IP address. The active gateway sends the encoded NAT record to the standby gateway.