diff options
| -rw-r--r-- | share/doc/papers/nqnfs/nqnfs.me | 12 |
1 files changed, 5 insertions, 7 deletions
diff --git a/share/doc/papers/nqnfs/nqnfs.me b/share/doc/papers/nqnfs/nqnfs.me index fba2a3b..9502ae1 100644 --- a/share/doc/papers/nqnfs/nqnfs.me +++ b/share/doc/papers/nqnfs/nqnfs.me @@ -33,6 +33,7 @@ .\" SUCH DAMAGE. .\" .\" @(#)nqnfs.me 8.1 (Berkeley) 4/20/94 +.\" $FreeBSD$ .\" .lp .nr PS 12 @@ -56,7 +57,10 @@ Association. .nr PS 12 .ps 12 .ce -.AB +\fBAbstract\fR +.nr PS 10 +.ps 10 +.pp There are some constraints inherent in the NFS\(tm\(mo protocol that result in performance limitations for high performance @@ -72,8 +76,6 @@ The emphasis is on observed performance of a preliminary implementation of the protocol, in order to show how well this design works and to suggest possible areas for further improvement. -.AE - .sh 1 "Introduction" .pp It has been observed that @@ -1086,7 +1088,6 @@ The results for the four variants are plotted in figures 2-5. .(z .ps -1 -.R .TS box, center; c s s s s s s s @@ -1549,7 +1550,6 @@ The \fINo Caching\fR cases does not do any client side caching, performing all system calls via synchronous RPCs to the server. .(z .ps -1 -.R .TS box, center; c s s s s s s @@ -1587,7 +1587,6 @@ elapsed running time and, as such, is the optimistic bound on performance achievable with an ideal client caching scheme that never blocks for I/O. .(z .ps -1 -.R .TS box, center; c s s s s s s s s s @@ -1752,7 +1751,6 @@ line dotted from 3.862,9.387 to 4.600,10.200 .)z .(z .ps -1 -.R .TS box, center; c s s s s s s s s |
