NonStop Systems Introduction

NonStop Server Architecture
NonStop Systems Introduction—527825-001
7-12
Easy Expansion of Data
By contrast, adding another processor in a shared-memory configuration does not
result in a commensurate increase in processing power, because all the processors
are contending for the same memory resources.
Figure 7-8 on page 7-12 shows that the performance of a NonStop server tends to
increase at a linear rate as more processors are added.
Easy Expansion of Data
You have seen that the NonStop server accommodates growing use of an application
or set of applications by expanding in cost-effective increments. You just add as many
processors as you need to handle the increasing numbers of workstations on the
system.
But the NonStop server must also be able to handle another type of unpredictable
growth: the increasing amount of data that an application or set of applications might
need to process.
For example, if the area served by a gas and electric utility experiences rapid growth in
population, the utility must add thousands of customer records to its customer
Figure 7-8. Linear Scalability of NonStop Servers
8
7
6
5
4
3
2
1
0
Relative performance
1 2 3 4 5 6 7 8
VST097.vsd
Number of processors