Information Technology Reference
In-Depth Information
80
64
48
32
16
0
2
4
6
8
10
12
Number of Servers
Concurrent Push
Concurrent Push w/AGSS
Concurrent Push w/AGSS & SSS
Staggered Push
Figure 12.9
Server buffer requirement versus system scale
can be supported. Hence, server buffer requirement will not become a limiting factor to the
system's scalability.
Using the same parameters, the client buffer requirement is calculated to be 256KB. This
translates into an average prefill delay of 1.41 seconds. To determine the system response time,
we assume that the system is at 90% utilization. Then the corresponding scheduling delay
will be 0.735 seconds. Together with prefill delay, the average system response time becomes
2.146 seconds, well within acceptable limits. We perform more detailed sensitivity analysis
with respect to key system parameters in the following sections.
12.6.2 Server Buffer Requirement
Figure 12.9 plots server buffer requirement versus system scale (i.e., number of servers) for
both concurrent push and staggered push. This graph clearly shows the remarkable property of
staggered push - constant server buffer requirement irrespective of system scale. By contrast,
server buffer requirement increases with system scale under concurrent push, even with AGSS
and SSS. When concurrent push is scaled up to 12 servers, server buffer requirement increases
to 40.6MB compared to just 6.25MB under staggered push. Hence the ultimate scalability of
the concurrent push architecture will be limited by server buffer, while the proposed staggered-
push architecture can be scaled up without any upgrade to the existing servers.
12.6.3 Client Buffer Requirement
Figure 12.10 plots client buffer requirement versus system scale for both concurrent push and
staggered push. We observe that concurrent push is not scalable without SSS, while staggered
Search WWH ::
Custom Search