Microsoft Corporation commissioned eTesting Labs to compare
the scalability of Exchange 2000 running under Microsoft Windows 2000 Advanced
Server to the scalability of Microsoft Exchange Server 5.5 running under
Microsoft Windows NT 4.0 Enterprise Edition. Microsoft requested that eTesting
Labs perform this comparison by using an appropriate test tool to simulate the
real-world usage of a large population of simulated Post Office Protocol (POP)
and Simple Mail Transfer Protocol (SMTP) users. eTesting Labs chose the Ziff
Davis E-mail Server Test Tool (ZDESTT) for this task.
The results of
this test, which tested large numbers of simulated users, showed a significant
scalability and performance increase from Exchange Server 5.5 to Exchange 2000.
Specifically, eTesting Labs found the following results:
- Exchange 2000 was 66 percent more scalable than Exchange
Server 5.5. Exchange 2000 delivered e-mail for 66 percent more simulated users
than Exchange Server 5.5. On the four processor servers that were used,
Exchange 2000 processed the message traffic for up to 83,000 simulated users,
whereas Exchange Server 5.5 handled only up to 50,000 simulated users before it
reached its capacity.
- Exchange 2000 latency performance exceeded Exchange Server
5.5 by more than 20 percent. At all of the simulated user load levels that were
tested, Exchange 2000 produced lower average server response times compared to
Exchange Server 5.5 (by at least 20 percent).
See the "Test Results" section and the "Test Methodology"
section of this white paper for complete details.