This document describes the current supported platforms for Cisco Personal Assistant 1.4.
HARDWARE
The following current shipping platforms are supported for single-language use and multiple-language use of Cisco Personal Assistant 1.4:
• Cisco Media Convergence Server (MCS) 7835-H1-Features a single Intel Xeon 3.4-Ghz processor.
• Cisco Media Convergence Server (MCS) 7835-I1-Features a single Intel Xeon 3.4-Ghz processor.
• Cisco Media Convergence Server (MCS) 7825H-3000-Features a single Intel Pentium 4 3.06-Ghz processor.
• Cisco Media Convergence Server (MCS) 7825I-3000-Features a single Intel Pentium 4 3.06-Ghz
• HP DL 380G-4
• IBM x346
• HP DL 320G-2
• IBM x306
You can find descriptions of these platforms at Cisco 7800 Series Media Convergence Servers
The following platforms are also supported for single-language use and multiple-language use of Cisco Personal Assistant 1.4:
• Cisco Media Convergence Server (MCS) 7835H-3000
• Cisco Media Convergence Server (MCS) 7835I-3000
• Cisco Media Convergence Server (MCS) 7835H-2400
• Cisco Media Convergence Server (MCS) 78235I-2400
• Cisco Media Convergence Server (MCS) 7825H-2266
• HP DL 380G3
• IBM x345
• HP DL 320G2
The following platforms are also supported for single-language use or multiple-language use of Personal Assistant 1.4:
• Cisco Media Convergence Server (MCS) 7835-1000
• Cisco Media Convergence Server (MCS) 7825-800
• HP DL 380G2
• HP DL 320
These platforms are described in the release notes for Cisco Personal Assistant 1.3(4) on Cisco.com at Cisco Personal Assistant Release Notes for Cisco Personal Assistant Release 1.3(4)
The following platforms are also supported for single-language or multiple-language use of Personal Assistant 1.4:
• Cisco Media Convergence Server (MCS) 7835-1000
• Cisco Media Convergence Server (MCS) 7825-800
These platforms are described in the release notes for Cisco Personal Assistant 1.2(3) on Cisco.com at Cisco Personal Assistant Release Notes for Cisco Personal Assistant 1.2
SYSTEM CAPACITY
Cisco Media Convergence Servers (MCS) 7835-H1 or 7835-I1 and 7825H-3000 or 7825I-3000:
Table 1.
Server Installation
MCS-7825H-3000 or MCS-78251-3000
MCS-7835-H1 or MCS-7835-I1
Cisco Personal Assistant Users*
Speech Recognition Sessions
TTS Sessions
Cisco Personal Assistant Users*
Speech Recognition Sessions
TTS Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on different systems of the same type
2500 Users
60 Sessions
12 Sessions
2500 Users
88 Sessions
12 Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on the same system
2500 Users
48 Sessions
12 Sessions
2500 Users
52 Sessions
12 Sessions
*Cisco Personal Assistant supports 15,000 busy hour call attempts (BHCA). Divided by an estimated six calls per person per hour, approximately 2500 users can be supported. See the Cisco Personal Assistant Design Guide for detailed deployment scenarios.
Table 2. Using Acoustic Model English America
Server Installation
20,000 Names in Corporate Directory
45,000 Names in Corporate Directory
Cisco Personal Assistant Users*
Speech Recognition Sessions
TTS Sessions
Cisco Personal Assistant Users*
Speech Recognition Sessions
TTS Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on different systems of the same type
2500 Users
60 Sessions
12 Sessions
2500 Users
88 Sessions
12 Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on the same system
2500 Users
48 Sessions
12 Sessions
2500 Users
52 Sessions
12 Sessions
*Cisco Personal Assistant supports 15,000 busy hour call attempts (BHCA). Divided by an estimated six calls per person per hour, approximately 2500 users can be supported. See the Cisco Personal Assistant Design Guide for detailed deployment scenarios.
For acoustic models other than American English, expect a performance capacity range of the following:
• Approximately the same performance capacity for other locales with a similar acoustic model (plus or minus 5%).
• Approximately 5-10% reduced performance capacity for each additional locale running on the same PA speech server.
• Approximately 15% reduced performance if the maximum number of enhanced TTS sessions configured.
CISCO MEDIA CONVERGENCE SERVERS (MCS) 7835H-3000 OR 7835I-3000 AND 7825H-3000 SYSTEMS:
Table 3. Server System Capacity Using Acoustic Model English America.1, 10,000 Names in Corporate Directory
Server Installation
MCS-7825H-3000
MCS-7835H-3000 or MCS-7835I-3000
Cisco Personal Assistant Users*
Speech Recognition Sessions
TTS Sessions
Cisco Personal Assistant Users*
Speech Recognition Sessions
TTS Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on different systems of the same type
2500 Users
60 Sessions
12 Sessions
2500 Users
88 Sessions
12 Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on the same system
2500 Users
48 Sessions
12 Sessions
2500 Users
52 Sessions
12 Sessions
*Cisco Personal Assistant supports 15,000 busy hour call attempts (BHCA). Divided by an estimated six calls per person per hour, approximately 2500 users can be supported. See the Cisco Personal Assistant Design Guide for detailed deployment scenarios.
Table 4. Using Acoustic Model English America.3
Server Installation
20,000 Names in Corporate Directory
45,000 Names in Corporate Directory
Cisco Personal Assistant Users*
Speech Recognition Sessions
TTS Sessions
Cisco Personal Assistant Users*
Speech Recognition Sessions
TTS Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on different systems of the same type
2500 Users
60 Sessions
12 Sessions
2500 Users
88 Sessions
12 Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on the same system
2500 Users
48 Sessions
12 Sessions
2500 Users
52 Sessions
12 Sessions
*Cisco Personal Assistant supports 15,000 busy hour call attempts (BHCA). Divided by an estimated six calls per person per hour, approximately 2500 users can be supported. See the Cisco Personal Assistant Design Guide for detailed deployment scenarios.
For acoustic models other than American English, expect a performance capacity range of the following:
• Approximately the same performance capacity for other locales with a similar acoustic model (plus or minus 5%).
• Approximately 5-10% reduced performance capacity for each additional locale running on the same PA speech server.
• Approximately 15% reduced performance if the maximum number of enhanced TTS sessions configured.
CISCO MEDIA CONVERGENCE SERVERS (MCS) 7835H-2400 OR 7835I-2400 AND 7825H-2266
Table 5. Server System Capacity Using Acoustic Model English America.1, 10,000 Names in Corporate Directory
Server Installation
MCS-7825H-2266
MCS-7835H-2400 or MCS-7835I-2400
Cisco Personal Assistant Users*
Speech Recognition Sessions
TTS Sessions
Cisco Personal Assistant Users*
Speech Recognition Sessions
TTS Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on different systems of the same type
2500 Users
60 Sessions
12 Sessions
2500 Users
88 Sessions
12 Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on the same system
2500 Users
48 Sessions
12 Sessions
2500 Users
52 Sessions
12 Sessions
*Cisco Personal Assistant supports 15,000 busy hour call attempts (BHCA). Divided by an estimated six calls per person per hour, approximately 2500 users can be supported. See the Cisco Personal Assistant Design Guide for detailed deployment scenarios.
Table 6. Using Acoustic Model English America
Server Installation
20,000 Names in Corporate Directory
45,000 Names in Corporate Directory
Cisco Personal Assistant Users*
Speech Recognition Sessions
TTS Sessions
Cisco Personal Assistant Users*
Speech Recognition Sessions
TTS Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on different systems of the same type
2500 Users
60 Sessions
12 Sessions
2500 Users
88 Sessions
12 Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on the same system
2500 Users
48 Sessions
12 Sessions
2500 Users
52 Sessions
12 Sessions
*Cisco Personal Assistant supports 15,000 busy hour call attempts (BHCA). Divided by an estimated six calls per person per hour, approximately 2500 users can be supported. See the Cisco Personal Assistant Design Guide for detailed deployment scenarios.
For acoustic models other than American English, expect a performance capacity range of the following:
• Approximately the same performance capacity for other locales with a similar acoustic model (plus or minus 5%).
• Approximately 5-10% reduced performance capacity for each additional locale running on the same PA speech server.
• Approximately 15% reduced performance if the maximum number of enhanced TTS sessions configured.
CISCO MEDIA CONVERGENCE SERVERS (MCS) 7835-1266 AND 7825-1133
Table 7. Server System Capacity Using Acoustic Model English America.1, 10,000 Names in Corporate Directory
Server Installation
MCS-7825-1133
MCS-7835-1266
Cisco Personal Assistant Users*
Speech Recognition Sessions
Cisco Personal Assistant Users*
Speech Recognition Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on different systems of the same type
2500 Users
60 Sessions
2500 Users
88 Sessions
Cisco Personal Assistant server, speech recognition server, and enhanced TTS server installed on the same system
2500 Users
48 Sessions
2500 Users
52 Sessions
*Cisco Personal Assistant supports 15,000 busy hour call attempts (BHCA). Divided by an estimated six calls per person per hour, approximately 2500 users can be supported. See the Cisco Personal Assistant Design Guide for detailed deployment scenarios.
USING ACOUSTIC MODEL ENGLISH AMERICA.3
Table 8. Server System Capacity Using Acoustic Model English.America.3
Server Installation
20,000 Names in Corporate Directory
45,000 Names in Corporate Directory
Cisco Personal Assistant Users*
Speech Recognition Sessions
Cisco Personal Assistant Users*
Speech Recognition Sessions
Cisco Personal Assistant server and speech recognition server installed on different MCS-7835-1266 systems
2500 Users
60 Sessions
2500 Users
88 Sessions
Cisco Personal Assistant server and speech recognition server installed on the same MCS-7835-1266 system
2500 Users
48 Sessions
2500 Users
52 Sessions
*Cisco Personal Assistant supports 15,000 busy hour call attempts (BHCA). Divided by an estimated six calls per person per hour, approximately 2500 users can be supported. See the Cisco Personal Assistant Design Guide for detailed deployment scenarios.
For acoustic models other than American English, expect a performance capacity range of the following:
• Approximately the same performance capacity for other locales with a similar acoustic model (plus or minus 5%).
• Approximately 5-10% reduced performance capacity for each additional locale running on the same PA speech server.
• Approximately 15% reduced performance if the maximum number of enhanced TTS sessions configured.
