🚀 go-pugleaf

axs@cs.bham.ac.uk (Aaron Sloman) writes:

>This discussion omits another option: instead of character-cell
>terminals you can use X terminals connected to a powerful,
>expandable, shared compute/file server. For many kinds of users this
>combination provides most of the benefits of a workstation on each
>user's desk but at a lower cost per user (including memory costs,
>system management costs, etc.). (Obviously some kinds of users
>should not be on a shared machine, or should not have their screens
>driven over a network, e.g. when doing continuous high speed
>bit-mapped graphics.)
>
>There are now several computer science departments in universities
>where this network architecture works very well. Of course, it is
>very important that as usage grows you can increase the power of the
>shared machine, either by replacing the processor or by increasing
>the number of processors, if you have a decent symmetric
>multiprocessor system.
>
>Moreover, when there's a new step in technology it is generally
>cheaper to upgrade a shared central server, from which everyone
>immediately benefits, than to upgrade the workstation on every desk.
>X terminal users will generally experience a central server upgrade
>as comparable to replacing a desktop workstation with a more
>powerful model, or giving them all more memory, etc.
>
>People who suffered timeshared systems in the 70's often refuse to
>believe any of this unless they have direct experience of the
>benefits. (And people whose income depends on selling or looking
>after workstations won't believe it even then....)
>
>The strength of this argument is not undermined by the horrors of X:
>at least it provides graphical interaction over a network.
>
>Aaron

X-terminals are not themselves the cure-all either.  Like diskless
workstations they rely on decent networking as they produce quite a lot
of network traffic.  Ethernet with its CSMA/CD architecture and 10Mb/sec
bandwidth can be brought to its knees with these devices when the usage
is moderate or more (depending as well on the # of stations).  At
least with multiuser systems the terminal equipment was all character
based using I/O controllers that were often distributed (multipoint)
for the big guys or point-to-point between the terminal and the multiuser
computer systems.

The rule with ethernet one must remember is that just because you
have 10Mb/sec (1.25M bytes/sec RAW ABSOLUTE MAX THROUGHPUT without
taking into account any protocol overhead or collision degredation)
does not mean you can forget about well designed connectivity options.

The world of OLTP is a far different than academic/college environments.
I've been in both.

There is no one-solution-fits-all.  The best approach is to use the right
solution and implement a very supportive underlying networking
infrastructure.  Some of the older connectivity ideas still have their
place and are more suited to the job than some of the more modern ones.

--
-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Michael F. Santangelo                 + Internet: mike@cbl.umd.edu      [work]
Computer & Network Systems Head       +           mike@kavishar.umd.edu [home]
Univ MD: CEES / CBL (Solomons Island) + BITNET:   MIKE@UMUC  [fwd to mike@cbl]