You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 8 Next »

In this section:

 

Number of IP Addresses per Call Leg Size

The table below defines the number of

Unable to show "metadata-from": No such page "_space_variables"
call legs supported for a given number of IP Addresses assuming UDP ports 4,000 through 64,444 are configured for media port usage.

For 

Unable to show "metadata-from": No such page "_space_variables"
media resource allocation scenarios, the 
Unable to show "metadata-from": No such page "_space_variables"
requires a minimum of 10 IP addresses configured to support 150,000 call legs (one Primary IP address plus 9 alternate IP addresses in non-VLAN configuration).

Number of IP Addresses Needed to Support Call Legs

Number of Call LegsNumber of IP
Addresses
150,00010
140,0009
130,0008
120,0007
110,0006
96,0005
80,0004
64,0003
45,0002
25,0001

For example, to use 100,000 call legs you must configure at least six (6) IP addresses because five IP addresses will only allow up to 96,000 call legs through the

Unable to show "metadata-from": No such page "_space_variables"
.

The  

Unable to show "metadata-from": No such page "_space_variables"
supports configuring up to 14 alternate media addresses. For more details, see Alternate Media IP Addresses Support.

Configuration Limits

The 

Unable to show "metadata-from": No such page "_space_variables"
supports a large number of configurable objects and is capable of supporting from a few very large customers to thousands of small customers. Important details about the provisioning system include:

  • Interactive provisioning of configuration objects (such as commands typed at the CLI or entered in the EMA or EMS) does not affect call handling or capacity.
  • Provisioning behavior does not depend on the number of existing entities. The addition of the 10,000th instance is roughly comparable to adding the 1st such instance.
Caution

Bulk or non-interactive provisioning may affect call handling. Operations such as sourcing a CLI script with many configuration commands (or even a few commands affecting particularly large objects) should only be done during a maintenance window or at non-busy-hour intervals. 

Provisioning Limits

Unable to show "metadata-from": No such page "_space_variables"
provisioning limits are shown in the following table.

A minimum of 9 GB RAM is required to support the below SBC SWe configuration.

SBC Provisioning Limits

Configuration Objects

Provisioning Limits
 SBC 7000 SeriesSBC 5000 SeriesSBC SWe6
   VM 10-13 GB RAMVM 14+ GB RAM
Address Contexts4,0962,048128700
SIP Signaling Ports4,0962,048128700
H.323 Signaling Ports2,0482,048128128
SIP Trunk Groups

40,0001

10,0002

1281,400
H.323 Trunk Groups

2,0481

2,0482

128256
Gateway Trunks

1281

   
Zones4,0962,048128700
DNS Groups2,048512128128
IP Interfaces4,0964,0961282,048
IP Interfaces Groups4,0962,0481281,024
Alternate Media IP Addresses
(per LIF)
25425425452545
IP Addresses8,1924,096  
IP ACLs50,000316,0004 2,800
Dynamic Black List Entries4,096  128

Footnotes

1 - The aggregate number of IP trunk groups (SIP + GW + H.323) cannot exceed 40,000. As a best practice, do not provision more than 20,000 trunk groups in a single zone.

2 - The aggregate number of IP trunk groups (SIP + GW + H.323) cannot exceed 10,000.

3 - Includes approximately 40,000 IP ACLs for peers, 10,000 IP ACLs for signaling ports (2 x 4,000 SIP, 2,000 H323) and 100 IP ACLs for OAM.

4 - Includes approximately 10,000 IP ACLs for peers, 4,000 IP ACLs for signaling ports and 100 IP ACLs for OAM.

5 - SBC SWe supports a maximum of 64 alternate media IP addresses per vNIC.

6 - SBC SWe supports up to 1,000 G711 pass-through calls along with this configuration.

The time required to add an object to an existing configuration is immaterial (It takes much longer to actually type the command or complete the EMA/EMS fields). However, bulk loading of a large configuration from the CLI may take a substantial amount of time. 

Caution

See cautionary note above about the potential bulk configuration loading on call handling performance. 

While the 

Unable to show "metadata-from": No such page "_space_variables"
supports large limits on multiple types of configuration objects, a few typical usage scenarios exist. The length of time to bulk load a configuration depends on the size of the configuration. Consequently, four
Unable to show "metadata-from": No such page "_space_variables"
profile classes are defined to give customers a rough idea for the bulk loading time relevant to their configuration:

  • 1K profile – comprised of 1,000 instances each of Address Contexts, Zones, IP Interface Groups, IP Interfaces, Signaling Ports, and Trunk Groups
  • 4K profile – comprised of 4,000 instances each of Address Contexts, Zones, IP Interface Groups, IP Interfaces, Signaling Ports, and Trunk Groups
  • 4K/10K profile – comprised of 4,093 instances of IP Interfaces, 2,048 instances each of Zones, Address Contexts, IP Interface Groups, SIP Signaling Ports, 10,000 IP Peer, and 10,000 IP Trunks (see Note below).
  • 4K/40K profile [SBC 7000 series only] – comprised of 4,000 instances each of Address Contexts, Zones, IP Interface Groups, IP Interfaces, Signaling Ports; and 40,000 Trunk Groups
Note

If both IPv4 and IPv6 are combined on every sipSigPort on an SBC 5200, the system limits the number of sipSigPorts to 1,980.

Maximum CLI Load Times

The following table lists the maximum CLI load times for the above profile classes.

Maximum CLI Load Times

Configuration ProfileMaximum CLI Load Time
1K2.5 hours

4K

4 hours
4K/10K3 hours
4/40K (7000 only)10 hours

For a system previously loaded with configuration, the start-up time also depends on the amount of configuration. The start-up time is considered the time from when the

Unable to show "metadata-from": No such page "_space_variables"
application is started (by PM) to the time that the
Unable to show "metadata-from": No such page "_space_variables"
is ready to accept calls.

Note that this start-up time is typically only relevant to standalone systems after an upgrade. For a HA system, the standby is activated to handle calls while the SBC is being started or restarted. 

Approximate Start-up Times for Standalone SBC

The following table lists the approximate start-up times for a standalone

Unable to show "metadata-from": No such page "_space_variables"
based on its configuration profile:

Approximate Start-up Times for Standalone SBC

Configuration ProfileMaximum Time
to Service Ready
No configuration5 minutes
1K10 minutes
4K20 minutes
4K/10K22 minutes
4K/40K (7000 only)45 minutes