High Availability

Server Redundancy Options for Delivering High Availability

• Home Location Register redundancy (HLR) - In this case, you are deploying IP/SIP phones and have more than one server with load balancing enabled to share the load between servers. The user data is synchronized between the servers. Should one of the servers fail, the IP/SIP phones will re-register to one of the secondary servers (using a guest/backup HLR) automatically and recover most of their features. When the primary server comes back, they will re-register to their primary HLR server. Each server in an HLR redundancy setup can handle up to 15,000 “visitors” IP/SIP end-points.
• 1+1 server redundancy. This is a scenario where a primary server and a dedicated backup server are configured in a redundancy cluster. Should the primary server fail, the backup server will take over the service within less than a minute (in a Pre-Loaded Server deployment). This is an active standby setup. There can be a 1+1 redundancy cluster for each server in a MX-ONE logical system.
• N+1 server redundancy. This scenario is similar to the 1+1 scenario, except you have several servers in one location in the same redundancy cluster with a dedicated backup server that will take over if any one of the primary servers in the cluster fails. The recovery process in an N+1 configuration will take 2-3 minutes as the backup server must first reconfigure itself with the failed primary server configuration. This is a warm standby setup. Up to 10 servers can be in a single redundancy cluster. There is no limit to the number of redundancy clusters in a single logical system.
• High Availability provided via virtualization software (e.g. VMware HA/FT). The MX-ONE Service Node fully support VMware HA scenarios - both High Availability (Warm Standby) and Fault tolerance (Hot Standby) options. It assumes the proper virtualization infrastructure is put in place according to VMware specifications.

Network Redundancy

MX-ONE servers and Media Gateways support network redundancy using built-in NIC Bonding or Link failover mechanisms. MX-ONE Servers (ASU Series) come with 2 LAN 100/1GB LAN ports that can be configured for network bonding/failover based 802.3 AD to provide network redundancy. Additionally, the MGU based media gateways (Lite and Classic) have 2 LAN100/1GB LAN ports that can be configured for network redundancy using a link failover mechanism.

System Redundancy

An alternate system can take over tasks of a primary system suffering from, for example, power outage. This way the services in primary system can be moved to an alternate system. There can be one alternate system for one single MX-ONE logical system.

Branch Office Survivability

For Branch office situations where there are requirements for local PSTN access and survivability for IP/SIP terminals, sites can be equipped with a remote server and media gateway combination - known as a Survivable Branch Node (SBN). The SBN bundles include the necessary trunk, tie-line, and user licenses as well as necessary HW depending on the options or package selected. As it is using standard MX-ONE Service Node software, it can be configured and managed centrally from the MX-ONE Management suite as a remote networked node. The SBN software runs either in a standard 3U Lite gateway or alternatively in a compact 1U server/gateway appliance called the EX Controller.

Alternatively, small remote sites or branch offices can be equipped with a small cost-effective survivability gateway (GX Gateway) appliance for SIP and analogue phones, offering local public trunk access. This appliance is completely self-sufficient and in case of a loss of connectivity to the main site call manager, the GX gateway will take over the call control of the local phones and provide basic call phone services and local PSTN access until the connection to the central MX-ONE system is re-established.