lhybl@ncrcae.Columbia.NCR.COM (Larry Hybl) (10/20/90)
I am looking for information on network reliability and availability
as it relates to both end systems and intermediate systems.
Anything that prevents consumers and providers of network services
(clients/servers) from communicating is fair game for discussion.
I would appreciate hearing from anyone with data or ideas related
to the following topics.
General:
- methods for increasing network availability
- methods for reducing network dependence
(e.g. Remote Procedure Call (RPC) compilers that generate code to
run locally if the network or remote server is down. This would
be a degraded-mode option to eliminate dependence on a compute-server.)
Specific:
- prioritized causes for network non-availability
(the network itself - i.e. the "wire/cloud", failed network adapters,
buggy end-node network software, etc.)
- non-robust network software (e.g. can't use the switched telephone network
when the X.25 connection fails)
- ideas for minimizing MTTR, e.g. hot pluggable boards, hot pluggable software,
advances in net management, e.g. fault detection and correction.
- redundant wiring & adapters
- fault tolerant software, etc.
Regards and Thanks in advance.
--
Larry Hybl email: larry.hybl@Columbia.NCR.COM
NCR Corporation, E&M Columbia VOICE: 632-6282 / (803) 791-6282
3325 Platt Springs Rd. FAX: 632-6998 / (803) 791-6998
West Columbia, S.C. 29169