Sören Janstål

.   Tidigare gallups

Data Research DPU ab
Torsvikssvängen 34
181 34 Lidingö

Tel +46 (0)70 727 67 95
Skype sjanstal

eMail to Sören Janstål

Sören Janstål
Sören Janstål
Data Research DPU
Data Research DPU ab
Data Research DPU for Evaluation of Information Technology

Sören Janstål
Sören Janstål

Bloor Research Group:

The Realities of O/S Scalability

"...Vendors frequently take hold of and distort the meaning of benchmark results. Indeed,

it is entirely predictable that they will do so..."


See order form for price!



The Issues

When you wish to choose an operating system for your computing environment, there is very little authoritative work done on benchmarking in a comparative manner the major players in the market.  Vendors' own benchmarks are manipulated to the point of being useless, and there are bewildering claims as to whether there is adequate support, both in applications and drivers, for the system you wish to use.

Which operating system is best?  Obviously it depends on your requirements, your hardware and a host of other factors. Assuming you're running an RDBMS on a commercial server, will the operating system need to scale to two or more processors to keep up with the load put on it?  Can the file system and the network connections scale along with the number of users?

The Realities of O/S Scalability has been written to penetrate the vendors' marketing hype using a suite of performance tests which provides a quantitative comparison of the operating systems. The tests simulate the typical behaviour of an on-line transaction processing system running on a high-performance server, and we believe that performance has been benchmarked realistically to assist you in making selections based on real-life applications instead of laboratory results.


The Report

The Realities of O/S Scalability is about the scalability of operating systems. We have examined and benchmarked  the behaviour of four of the most popular operating systems (NetWare, OS/2, Solaris and Windows NT) and in doing so we are able to rate them against each other.

The suite of programs used in the testing, the ScaPA Suite, is explained, followed by a brief explanation of the system structure used to test the operating systems. The Operating System Choice provides the background context of computer hardware types and the operating systems which run on them. The Systems Under Test explains the choices of operating systems for this report as well as the hardware used to run them.

The System Level Tests are then explained, and, finally, the Results are graphed and explained. Special test software was developed in order to use the Pentium Model Specific Registers, which gives us a better knowledge of what is happening at a low level, and this is explained in Chapter 4, Test Software.


Products Evaluated in this Report

As explained earlier, the products evaluated here are the four operating systems, IBM OS/2 Warp, version 4 SMP, Microsoft Windows NT, version 4, Sun Solaris x86, version 2.5.1 and Novell NetWare, version 4.11. These operating systems were fitted to a Compaq ProLiant 4500 with up to 4 processors, a configuration which is relatively common in a mid-range application server environment for On Line Transaction Processing (OLTP) applications.

Tests were run on the Development Environment using a number of different compilers, the difference in the efficiency of which was unavoidable. The compilers used are listed below.

 Operating System  Compiler  Version
 IBM OS/2 Warp 4 Advanced SMP  IBM Visual Age C++  3.0
 Microsoft Windows NT4.0  Microsoft Visual C++  4.0
 Sun Solaris 2.5.1  Gnu C++  
 Novell NetWare 4.11  Watcom C++  10.6


Why buy this report?

Many claims in the market place for operating systems are bewildering for the buyer. Which is best for the type of applications you run? Can it scale effectively as your needs grow? If you are running major database applications, is the file system up to it.? Networks are essential in the corporate environment - how do the communication aspects of the operating system measure up?

If you are a buyer for a corporate IT department, the above questions will be uppermost in your mind. Bloor Research believes that this unique report will answer these questions in great detail and help you to come to a responsible and accurate decision as to which operating system to choose.

Whilst the nature of the tests, and the subject matter itself is very complex, the reader going through the whole report will have a comprehensive understanding of how these operating systems work, their strengths and weaknesses, and their suitability for the tasks envisaged.

by Ben Younger & Dr. Mike Norman

Related reports:

ERP market, just this minute

Look how the users rate their systems!

Give your rating of any system and get a market analyse report free of charge!

Give your rating of any IT vendor and get a market analyse report free of charge!

Data Research DPU
for Evaluation of Information Technology and Computing

[ Order | More info | Suggest new evaluations ]
[ Consulting | Price List | Mailing List ]
[ Contact ]

Back to Data Research DPU top page.

Data Research DPU ab - Torsvikssvngen 34, SE-181 34 Liding, Sweden - Tel +46 70 727 67 95 - Skype: sjanstal, SkypeIN: +46 8 559 25 900 Contact (email)


Space 5 for sponsor/e

Space 9 for sponsor/e


Changed 1999-09-21