Oracle's SPARC T4 servers set a new World Record for
Oracle's Siebel Platform Sizing and Performance Program
(PSPP) benchmark suite. The result used Oracle's
Siebel Customer Relationship Management (CRM) Industry Applications Release
8.1.1.4 and Oracle Database 11g
Release 2 running Oracle Solaris on three SPARC T4-2 and
two SPARC T4-1 servers.
The SPARC T4 servers running the Siebel PSPP 8.1.1.4 workload
which includes
Siebel Call Center and Order Management System demonstrates
impressive throughput performance of the SPARC T4 processor
by achieving 29,000 users.
This is the first Siebel PSPP 8.1.1.4 benchmark supporting 29,000
concurrent users with a rate of 239,748 Business Transactions/hour.
The benchmark demonstrates vertical and horizontal scalability of Siebel CRM Release 8.1.1.4 on SPARC T4 servers.
Performance Landscape
Systems
Txn/hr
Users
Call Center
Order Management
Response Times (sec)
1 x SPARC T4-1 (1 x SPARC T4 2.85 GHz) – Web
3 x SPARC T4-2 (2 x SPARC T4 2.85 GHz) – App/Gateway
1 x SPARC T4-1 (1 x SPARC T4 2.85 GHz) – DB
239,748
29,000
0.165
0.925
Oracle:
Call Center + Order Management
Transactions: 197,128 + 42,620
Users: 20300 + 8700
Configuration Summary
Web Server Configuration:
1 x SPARC T4-1 server
1 x SPARC T4 processor, 2.85 GHz
128 GB memory
Oracle Solaris 10 8/11
iPlanet Web Server 7
Application Server Configuration:
3 x SPARC T4-2 servers, each with
2 x SPARC T4 processor, 2.85 GHz
256 GB memory
3 x 300 GB SAS internal disks
Oracle Solaris 10 8/11
Siebel CRM 8.1.1.5 SIA
Database Server Configuration:
1 x SPARC T4-1 server
1 x SPARC T4 processor, 2.85 GHz
128 GB memory
Oracle Solaris 11 11/11
Oracle Database 11g Release 2 (11.2.0.2)
Storage Configuration:
1 x Sun Storage F5100 Flash Array
80 x 24 GB flash modules
Benchmark Description
Siebel 8.1 PSPP benchmark includes Call Center and Order Management:
Siebel Financial Services Call Center – Provides the most complete
solution for sales and service, allowing customer service and telesales
representatives to provide superior customer support, improve customer
loyalty, and increase revenues through cross-selling and up-selling.
High-level description of the use cases tested: Incoming Call Creates Opportunity, Quote and Order and Incoming Call Creates Service Request
. Three complex business transactions are executed simultaneously for specific number of concurrent users. The ratios of these 3 scenarios were 30%, 40%, 30% respectively, which together were totaling 70% of all transactions simulated in this benchmark. Between each user operation and the next one, the think time averaged approximately 10, 13, and 35 seconds respectively.
Siebel Order Management – Oracle's Siebel Order Management allows
employees such as salespeople and call center agents to create and
manage quotes and orders through their entire life cycle. Siebel Order
Management can be tightly integrated with back-office applications
allowing users to perform tasks such as checking credit, confirming
availability, and monitoring the fulfillment process.
High-level description of the use cases tested: Order & Order Items Creation and Order Updates. Two complex Order Management transactions were executed simultaneously for specific number of concurrent users concurrently with aforementioned three Call Center scenarios above. The ratio of these 2 scenarios was 50% each, which together were totaling 30% of all transactions simulated in this benchmark. Between each user operation and the next one, the think time averaged approximately 20 and 67 seconds respectively.
Key Points and Best Practices
No processor cores or cache were activated or deactivated on the
SPARC T-Series systems to
achieve special benchmark effects.
See Also
Siebel White Papers
SPARC T4-1 Server
oracle.com
OTN
SPARC T4-2 Server
oracle.com
OTN
Siebel CRM
oracle.com OTN
Oracle Solaris
oracle.com
OTN
Oracle Database 11g Release 2 Enterprise Edition
oracle.com
OTN
Disclosure Statement
Copyright 2012, Oracle and/or its affiliates. All rights reserved.
Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names
may be trademarks of their respective owners. Results as of 30 September 2012.