6-4
Device Manager Guide, Cisco ACE 4700 Series Application Control Engine Appliance
OL-26645-02
Chapter 6 Configuring Real Servers and Server Farms
Server Load Balancing Overview
The ACE appliance uses traffic classification maps (class maps) within policy maps to filter out
interesting traffic and to apply specific actions to that traffic based on the SLB configuration. You use
class maps to configure a virtual server address and definition.
If a primary real server fails, the ACE appliance takes that server out of service and no longer includes
it in load-balancing decisions. If you configured a backup server for the real server that failed, the ACE
appliance redirects the primary real server connections to the backup server. For information about
configuring a backup server, see the Configuring Virtual Server Layer 7 Load Balancing, page 5-30.
The ACE appliance can take a real server out of service for the following reasons:
• Probe failure
• ARP timeout
• Neighbor Discovery (ND) failure (IPv6 only)
• Retcode failure
• Reaching the maximum number of connections
• Specifying Out Of Service as the administrative state of a real server
• Specifying In Service Standby as the administrative state of a real server
The Out Of Service and In Service Standby selections both provide the graceful shutdown of a server.
Related Topics
• Configuring Real Servers, page 6-5
• Configuring Health Monitoring for Real Servers, page 6-41
Dynamic Workload Scaling Overview
The ACE Dynamic Workload Scaling feature permits on-demand access to remote resources, such as
VMs, that you own or lease from an Internet service provider or cloud service provider. This feature uses
Cisco Nexus 7000 Series switches with Overlay Transport Virtualization (OTV) technology to create a
Data Center Interconnect (DCI) on a Layer 2 link over an existing IP network between geographically
distributed data centers. The local data center Nexus 7000 contains an OTV forwarding table that lists
the MAC addresses of the Layer 2 extended virtual private network (VPN) and identifies the addresses
as either local or remote.
When you configure the ACE to use this feature, the ACE uses an XML query to poll the Cisco Nexus
7000 Series Switch and obtain the OTV forwarding table information to determine the locality of the
local or remote VMs. The ACE also uses a health monitor probe that it sends to the local VMware
vCenter Server to monitor the load of the local VMs based on CPU usage, memory usage, or both. When
the average CPU or memory usage of the local VMs reaches its configured maximum threshold value,
the ACE bursts traffic to the remote VMs. The ACE stops bursting traffic to the remote VMs when the
average CPU or memory usage of the local VMs drops below its configured minimum threshold value.
To use Dynamic Workload Scaling, you configure the ACE to connect to the Data Center Interconnect
device (Cisco Nexus 7000 Series switch) and the VMware Controller associated with the local and
remote VMs. You also configure the ACE with the probe type VM to monitor a server farm’s local VM
CPU and memory usage, which determines when the ACE bursts traffic to the remote VMs.
For more details on this feature, see the Server Load-Balancing Guide, Cisco ACE Application Control
Engine.
Comments to this Manuals