IMP System change notification
An announcement of IMP system version 2505 with expanded capacity and improved monitoring
Cosell, B., "IMP System change notification", RFC 213, DOI 10.17487/RFC0213, August 1971, <https://www.rfc-editor.org/info/rfc213>.
@misc{rfc213,
series = {Request for Comments},
number = 213,
howpublished = {RFC 213},
publisher = {RFC Editor},
doi = {10.17487/RFC0213},
url = {https://www.rfc-editor.org/info/rfc213},
author = {B. Cosell},
title = {{IMP System change notification}},
pagetotal = 1,
year = 1971,
month = aug,
} RFC text
Network Working Group B. Cosell
Request for Comments: #213 BBN
NIC #7194 20 August 1971
Categories: B.1
Updates: None
Obsoletes: None
IMP System Change Notification
------------------------------
During the week of 16 August, we installed a new version of
the IMP system, version 2505. The new version incorporated three
principal changes:
1) The number of hosts the system can handle was increased
from three to four. Although this caused a rather large
internal change to the IMP, the only external effects of
the change are the presence of a new status light [light
8 monitors Host 4], and the lengthening of the 10-second
and snapshot statistics.
2) The host handling routines were modified to gather more
accurate and voluminous statistics on host usage of the
network, and report them to NCC.
3) The 10-sec. statistics were changed to measure time for
the host "round trip" statistic in units of 800 rather
than 100 microseconds.
It is not anticipated that the new system should have any
direct effect on the hosts.
[ This RFC was put into machine readable form for entry ]
[ into the online RFC archives by BBN Corp. under the ]
[ direction of Alex McKenzie. 12/96 ]
AI ELI5
On this page
§1The Update
During the week of 16 August 1971, BBN installed a new version of the IMP system, version 2505. This update incorporated three principal changes to improve network capacity and monitoring.
The primary change was expanding the number of hosts the system could handle from three to four. Although this required significant internal changes to the IMP's host-handling routines, the only external effects were the presence of a new status light (light 8 monitors Host 4) and the lengthening of the 10-second and snapshot statistics reports.
An upgrade added room for one more computer to connect, with only a new indicator light to show the change on the outside.
§2Improved Statistics
The host-handling routines were enhanced to gather more accurate and voluminous statistics on host usage of the network and report them to the Network Control Center. This gave the NCC better insight into how hosts were using network resources.
Additionally, the 10-second statistics were changed to measure the host "round trip" statistic in units of 800 microseconds rather than 100 microseconds. This change provided more precise timing measurements while consuming less space in the statistics reports.
The system now keeps better track of what each computer is doing and measures time more precisely.
§3Impact on Hosts
BBN anticipated that the new system should have no direct effect on the hosts themselves. The expansion and improvements were internal to the IMP infrastructure and should be transparent to host software and users. Hosts would continue to operate normally with the same interface and behavior.
Everything works the same from the host's point of view. The changes are all behind the scenes.