.(z C
.(c
.TS
center, tab(;);
| c | c s | c s | c s |
| c | c s | c s | c s |
| c | c | c | c | c | c | c |
| c | c | c | c | c | c | c |
| n | n | n | n | n | n | n |.
.br
_
Benchmark;Degradation;Server Utilization;Network Utilization
\^;_;_;_
\^;Original;Handle;Original;Handle;Original;Handle
\^;\^;Locally;\^;Locally;\^;Locally
=
Andrew;12.5%;6.2%;8.9%;5.6%;1.04%;.69%
_
Fs-make;5.6%;1.0%;6.7%;4.1%;.75%;.45%
_
.TE
.)c
.(q
\fBTable 8\fR. Estimated improvements possible from client-level name
caching and server callback.  The estimates were made by counting
invocations of Open and Get Attributes operations in the benchmarks
and recalculating degradations and utilizations under the assumption
that all of these operations could be executed by clients without any
network traffic or server involvement.
.)q
.)z