OpenNetwork Technologies sponsored the testing in
this report. Mindcraft, Inc. conducted the performance tests described
in this report at Sun’s test lab in Frankfurt, Germany.
We thank Sun for providing the systems used for the tests and the
support staff who helped configure the servers.
DirectorySmart 4.6 delivers outstanding
performance scaling and achieves the highest login and Extranet
sequence rates we've seen to date: 146,051 logins per minute
and 25,428 Extranet Sequences (279,708 operations) per minute.
Mindcraft® tested OpenNetwork Technologies DirectorySmart
4.6 running on Sun Enterprise servers. For these tests, we used
Mindcraft’s iLOAD MVP™ test
tool running the AuthMark™
Login and Extranet
Scenarios. During these tests DirectorySmart set new performance
records for authentication/authorization products while providing
almost linear performance scaling.
The Login Scenario represents the type of load commonly seen at
portal sites. It simulates users accessing protected resources via
Web servers. The Login Scenario assumes that 10% of a portal's user
population logs in concurrently to use portal resources. All
tests were done using a 1,000,000-user directory with 100,000 active
The Login Scenario measures the combination of one user authentication
and one authorization for access to a resource (called a Login).
The Result Analysis section
in the second part of this white
paper explains the benchmark results.
DirectorySmart, which is located on a Web server for the configurations
we tested, is the control point for all authentication and authorization.
Our tests were structured to push the Web server systems as closely
as possible to 100% CPU utilization. DirectorySmart uses LDAP directory
servers to store user authentication and authorization information
without an intervening policy/authentication server. So, the performance
of DirectorySmart is closely coupled to that of the LDAP directory
servers. That is why Table 1 summarizes
the Login Scenario performance as a function of the Web and LDAP
server configurations. The Scaling Factor in Table 1 shows how much
faster a configuration is compared to the smallest configuration,
Table 1: DirectorySmart Login Performance
Scalability - 1,000,000-User Directory
The Web server CPU utilizations for Configurations 3, 5, and 6
show that more performance could have been derived from DirectorySmart.
The limiting factor in these cases was the performance of the load
Figure 1 shows DirectorySmart's Login
performance from Table 1 by server configuration. The following
color coding is used to group performance results by LDAP directory
- Purple: Two LDAP directory servers with one CPU each.
- Blue: Two LDAP directory servers with two CPUs each.
- Orange: Three LDAP directory servers with four CPUs each.
Look at the hardware configurations
in the second part of this report for more details on the test environment.
DirectorySmart Login Scalability for a 1,000,000-User Directory
The Extranet Scenario measures the combination of one user authentication
and 10 authorizations for access to resources (these 11 operations
constitute one Extranet sequence). The Extranet Scenario, because
it uses a more realistic mix of operations than the Login Scenario,
provides a better basis for capacity planning purposes.
Table 2 shows the DirectorySmart Extranet
Scenario performance for Configuration 6 in Table
1 - eight Web servers with four CPUs each and three LDAP directory
servers with four CPUs each. The results demonstrate that DirectorySmart
performs authorizations faster than it does authentications. The
CPU utilizations show that DirectorySmart could have achieved higher
performance; the load generator systems were running at 100% CPU
utilization, which limited our ability to drive the Web servers
with DirectorySmart to their maximum performance. The 20% CPU utilization
on the LDAP directory servers means that DirectorySmart uses them
much less for authorizations than it does for logins. This means
that you can plan to deploy more Web servers per LDAP directory
server than we used for these tests and can expect to achieve higher
authorization rates than we did.
Table 2: DirectorySmart Extranet Performance
- 1,000,000 User Directory
The benchmark results lead us to conclude that:
- OpenNetwork Technologies's DirectorySmart 4.6
has achieved the highest AuthMark Login and Extranet Scenario
performance we've seen to date.
- DirectorySmart 4.6 delivers very consistent
login performance per CPU, which makes it easy to plan configurations
for the load you need to handle.
- DirectorySmart delivers outstanding performance scaling as CPUs
are added to a configuration.
Mindcraft certifies that the results reported accurately represent
the performance of OpenNetwork Technologies's DirectorySmart 4.6
running on Sun Enterprise servers configured as specified herein
and as measured by AuthMark benchmark.
Our test results should be reproducible by others using the same
test lab configuration, the same Sun server configurations, and
the same software configurations documented in this white paper.
The information in this publication is subject to
change without notice.
MINDCRAFT, INC. SHALL NOT BE LIABLE FOR ERRORS OR OMISSIONS CONTAINED
HEREIN, NOR FOR INCIDENTAL OR CONSEQUENTIAL DAMAGES RESULTING FROM
THE FURNISHING, PERFORMANCE, OR USE OF THIS MATERIAL.
This publication does not constitute an endorsement
of the product or products that were tested. This test is not a
determination of product quality or correctness, nor does it ensure
compliance with any federal, state or local requirements.