Difference between revisions of "Host-to-IMP Protocol"
(Link to BBN.) |
m (→Physical layer: clarify) |
||
Line 9: | Line 9: | ||
At the very lowest layer, there are three different choices for the [[physical layer]]; 'Local Host' (LH), 'Distant Host' (DH), and 'Very Distant Host' (VDH). | At the very lowest layer, there are three different choices for the [[physical layer]]; 'Local Host' (LH), 'Distant Host' (DH), and 'Very Distant Host' (VDH). | ||
− | The first two are logically the same, and differ only in the physical details - DH uses [[transistor-transistor logic|TTL]] signals over cable using [[twisted pair]]s, with a [[ground]] being the other | + | The first two are logically the same, and differ only in the physical details - DH uses [[transistor-transistor logic|TTL]] signals over cable using [[twisted pair]]s, with a [[ground]] being the other [[conductor]] in the pair, whereas LH uses [[differential pair]]s. |
Above that, the interface is completely identical; it is [[asynchronous]], bit-serial, with a handshake on every bit. There is no fixed [[word]] length, the IMP was prepared to talk to machines with different [[word]] lengths. There are also two pairs of 'ready' lines (one from the IMP, and one from the Host), which are connected to ground via a [[relay]] when that entity wants to signal that it is ready. | Above that, the interface is completely identical; it is [[asynchronous]], bit-serial, with a handshake on every bit. There is no fixed [[word]] length, the IMP was prepared to talk to machines with different [[word]] lengths. There are also two pairs of 'ready' lines (one from the IMP, and one from the Host), which are connected to ground via a [[relay]] when that entity wants to signal that it is ready. |
Revision as of 13:45, 18 October 2021
The Host-to-IMP Protocol includes the lowest several layers (Physical, etc) of the communication channel between a host and an IMP. It used both for the host to communicate with its IMP, and for one host to send a packet to another, as the lowest layer of NCP.
The service provided was reliable (although if a packet was irretrievably lost, the IMP would indicate that to the host with an error message), with flow control.
Contents
Physical layer
Connection of a host to an IMP required a special network interface, an IMP interface.
At the very lowest layer, there are three different choices for the physical layer; 'Local Host' (LH), 'Distant Host' (DH), and 'Very Distant Host' (VDH).
The first two are logically the same, and differ only in the physical details - DH uses TTL signals over cable using twisted pairs, with a ground being the other conductor in the pair, whereas LH uses differential pairs.
Above that, the interface is completely identical; it is asynchronous, bit-serial, with a handshake on every bit. There is no fixed word length, the IMP was prepared to talk to machines with different word lengths. There are also two pairs of 'ready' lines (one from the IMP, and one from the Host), which are connected to ground via a relay when that entity wants to signal that it is ready.
The VDH uses an entirely different protocol, essentially the modem to modem protocol which the IMPs use to talk to each other over a long-distance synchronous serial line.
Network layer
Communication and between IMPs and their connected hosts was all done with packets (technically, "messages").
Host addresses
Host addresses in the ARPANET consisted of the IMP number, and the host number on the IMP. In the original 'short leader' format (below), IMP numbers were 6 bits, and host numbers were 2 bits. In the later 'long leader' format, these were increased to 16 bits and 8 bits respectively.
Packet formats
There were two versions of the Host-IMP headers:
- The so-called "short leader" format, used from 1970 until roughly the end of 1975 (the exact date is not yet ascertained), although it continued to be supported after that date, but hosts using it could not reach all of the other hosts in the network at some point shortly thereafter. It was 32 bits long.
- The so-called "long leader" format, introduced around the end of 1975, which allowed i) more IMPs in the network, ii) more hosts per IMP, and iii) various other improvements. It was 96 bits long.
For details, see the external link below.
Links
In both the original 32-bit leader and the later 96-bit leader, there was a 12-bit field called the 'message ID'; the high 8 bits of this field were designated as the link.
The NCP protocol used links 0, and 2-0107; the low 4 bits of the message ID were always set to zero. The Internet Protocol used links 0233-0236. Several other internetworking layer protocols had links assigned, to allow them to be carried over the ARPANET: 0230 for PUP, and 0227 for CHAOS.
Packet Types (Host->IMP)
The network layer uses a number of different packet types; some are used by the IMP and Host to communicate with each other, others are used to send data from one Host to another.
One important type is the 'Ready For Next Message' (RFNM), used between the Host and the IMP (with cooperation from the destination Host's IMP) for flow control; it indicated when a message had been successfully delivered to the destination Host, and that Host's IMP was ready for another packet to that host.
(Depending on the exact version of the code running in the IMP, there was generally a maximum of 8 outstanding packets to any given destination allowed at a time.)
Here are the packet types, listed by type and (indented) sub-type:
- 0 - Regular Message
- 0 - Standard, Non-Refusable
- 1* - Refusable
- 2* - Get Ready
- 3 - Uncontrolled
- 1 - Error Without Message Identification
- 0 - Host's error flip-flop set during transmission
- 1 - Host received a message less than 80 bits
- 2 - Host received a message of an unassigned type
- 2 - Host Going Down
- 0 - Reserved for IMP use
- 1 - Reserved for IMP use
- 2 - Reserved for IMP use
- 3 - Reserved for IMP use
- 4 - Reserved for IMP use
- 5 - Scheduled PM
- 6 - Scheduled hardware work
- 7 - Scheduled software work
- 8 - Reserved for IMP use
- 9 - Reserved for IMP use
- 10 - Emergency Restart
- 11 - Power Outage
- 12 - Software Breakpoint
- 13 - Hardware Failure
- 14 - Not scheduled to be up
- 15 - Unspecified Reason
- 3 - Unassigned
- 4 - NOP
- 5 - Unassigned
- 6 - Unassigned
- 7 - Unassigned
- 8 - Error with Message Identification
* = Non-blocking interface, not implemented
Packet Types (IMP->Host)
Listed by type and (indented) sub-type:
- 0 - Regular Message
- 1 - Error in Leader
- 0 - IMP's error flip-flop set during transmission
- 1 - IMP received a message less than 80 bits
- 2 - IMP received a message of an unassigned type
- 3 - IMP received a message of the opposite leader style than it was expecting
- 2 - IMP Going Down
- 3 - Unused
- 4 - NOP
- 5 - RFNM
- 0 - Original Message Nno-Refusable
- 1 - Original Message Refusable
- 6 - Dead Host Status
- 0 - Currently Unused
- 1 - Destination Host not communicating - took its ready-line down
- 2 - Destination Host not communicating - tardy taking traffic
- 3 - Destination Host does not exist
- 4 - IMP software is preventing communication with this Host
- 5 - Destination Host is down for scheduled PM
- 6 - Destination Host is down for scheduled hardware work
- 7 - Destination Host is down for scheduled software work
- 8 - Destination Host is down for emergency restart
- 9 - Destination Host is down because of a power outage
- 10 - Destination Host is stopped at a software breakpoint
- 11 - Destination Host is down because of a hardware failure
- 12 - Destination Host is not scheduled to be up
- 13 - Currently Unused
- 14 - Currently Unused
- 15 - Destination Host is in the process of coming up
- 7 - Destination Host or IMP Dead or Unknown
- 0 - The destination IMP cannot be reached
- 1 - The destination Host is not up
- 2 - Communication with the destination Host is not possible because it does not support new leaders
- 3 - Communication with the destination Host is administratively prohibited
- 8 - Error in Data
- 9 - Incomplete Transmission
- 0 - Destination Host did not accept the message quickly enough
- 1 - Message was too long
- 2 - The Host took too long to transmit the message to the IMP
- 3 - Message lost in the network dur to IMP or circuit failure
- 4 - The IMP could not accept the message quickly enough because of unavailable resources
- 5 - Source IMP I/O failure during receipt of this message
- 10 - Interface Reset
- 11* - Refused, Try Again
- 0 - IMP buffer was not available
- 1 - Transmit block for connection was not available
- 12* - Refused, Will Notify
- 0 - Currently unused
- 1 - Currently unused
- 2 - Connection not available
- 3 - Reassemly space not available at destination
- 4 - Message number not available
- 5 - Transaction block for message not available
- 13* - Refused, Still Trying
- 14* - Ready
* = Non-blocking interface, not implemented
See also
- BBN Report #1822, "Specification for the Inteconnection of a Host and an IMP" - The 1976 revision of this key document, it includes the earliest specification of the new long leader format