The most commonly run of these is the SAP-SD (Sales and Distribution) benchmark. It can be run in a
2-tier environment, where the application and database reside on the same system, or on a 3-tier
environment, where there are many application servers feeding into a database server.
Care must be taken to ensure that the same level of software is being run when comparing results of SAP
benchmarks. Like most software suppliers, SAP strives to enhance their product with useful functions in
each release. This can yield significantly different performance characteristics between releases such as
4.0B, 4.5B, and 4.6C. It should be noted that, although SAP is used as an example here, this situation is
not restricted to SAP software.
For more information on SAP benchmarks, go to
http://www.sap.com
and process a search for
Standard
Application Benchmarks Published Results
.
NotesBench
There are several benchmarks that are called "Notesbench xxx". All come from the Notesbench
Consortium, a consortium of vendors interested in using benchmarks to help quantify system capabilities
using Lotus Domino functions. The most popular benchmark is Notesbench R5 Mail, which is actually a
mail and calendar benchmark that was designed around the functions of Lotus Domino Release 5.0.
AS/400 and iSeries systems have traditionally demonstrated very strong performance in both capacity and
response time in Notesbench results.
For official iSeries audited NotesBench results, see
http://www.notesbench.org
. (Note: in order to access
the NotesBench results you will need to apply for a userid/password through the Notesbench
organization. Click on Site Registration at the above address.) An alternate is to refer to the
ideasInternational web site listed above.
For more information on iSeries performance in Lotus Domino environments, refer to Chapter 11 of this
document.
SPECjbb2000
The Standard Performance Evaluation Corporation (SPEC) defined, in June, 2000, a server-side Java
benchmark called SPECjbb2000. It is one of the only Java-related benchmarks in the industry that
concentrates on activity in the server, rather than a single client. The iSeries architecture is well suited for
an object-oriented environment and it provides one of the most efficient and scalable environments for
server-side Java workloads. iSeries and AS/400 results are consistently at or near the top rankings for this
benchmark.
For more information on SPECjbb2000 and for published results, see
http://www.spec.org/osg/jbb2000/
For more information on iSeries performance in Java environments, refer to Chapter 7 of this document.
VolanoMark
IBM has chosen the VolanoMark benchmark as another means for demonstrating strength with
server-side Java applications. VolanoMark is a 100% Pure Java server benchmark characterized by
long-lasting network connections and high thread counts. It is as much a test of tcp/ip strengths as it is of
multithreaded, server-side Java strengths. In order to scale well in this benchmark, a solution needs to
scale well in tcp/ip, Java-based applications, multithreaded application, and the operating system in
general. Additional information on the benchmark can be found at
http://www.volano.com/benchmarks.html
.
IBM i 6.1 Performance Capabilities Reference - January/April/October 2008
©
Copyright IBM Corp. 2008
Chapter 19 - Misc Perf Information
302
