The processing of workunits and results
can be described in terms of transitions of their state variables.
Workunit state variables
Workunits parameters are described here.
Workunit state variables are as follows:
";
list_start();
list_item(
"canonical_resultid",
"The ID of the canonical result for this workunit, or zero.
- Initially zero
- Set by the validator (by check_set())
"
);
list_item("transition_time",
"The next time to check for state transitions for this WU.
- Initially now.
- Set to now by scheduler when get a result for this WU.
- Set to min(current value, now + delay_bound) by scheduler
when send a result for this WU
- Set to min(x.sent_time + wu.delay_bound) over IN_PROGRESS results x
by transitioner when done handling this WU
- Set to now by validator if it finds canonical result,
or if there is already a canonical result
and some other results have validate_state = INIT,
or if there is no consensus and the number of successful results
is > wu.max_success_results
"
);
list_item("file_delete_state",
"Indicates whether input files should be deleted.
- Initially INIT
- Set to READY by transitioner when all results have server_state=OVER
and wu.assimilate_state=DONE
Note: db_purge purges a WU and all its results when
file_delete_state=DONE;
therefore it is critical that it only be set to DONE
if all results have server_state=OVER.
- Set to DONE by file_deleter when it has attempted to delete files.
"
);
list_item("assimilate_state",
"Indicates whether the workunit should be assimilated.
- Initially INIT
- Set to READY by transitioner if wu.assimilate_state=INIT
and WU has error condition
- Set to READY by validator when find canonical result
and wu.assimilate_state=INIT
- Set to DONE by assimilator when done
"
);
list_item("need_validate",
"Indicates that the workunit has a result that needs validation.
- Initially FALSE
- Set to TRUE by transitioner if the number of success results
is at least wu.min_quorum and there is a success result
not validated yet
- Set to FALSE by validator
"
);
list_item("error_mask",
"A bit mask for error conditions.
- Initially zero
- Transitioner sets COULDNT_SEND_RESULT if some result couldn't be sent.
- Transitioner sets TOO_MANY_RESULTS if too many error results
- Transitioner sets TOO_MANY_TOTAL_RESULTS if too many total results
- Validater sets TOO_MANY_SUCCESS_RESULTS if no consensus
and too many success results
"
);
list_end();
echo "
Workunit invariants:
- eventually either canonical_resultid or error_mask is set
- eventually transition_time = infinity
- Each WU is assimilated exactly once
Notes on deletion of input files:
- Input files are eventually deleted,
but only when all results have state=OVER
(so that clients don't get download failures)
and the WU has been assimilated
(in case the project wants to examine input files in error cases).
Result state variable
Result state variables are listed in the following table:
";
list_start();
list_item("report_deadline",
"Give up on result (and possibly delete input files)
if don't get reply by this time.
- Set by scheduler to now + wu.delay_bound when send result
"
);
list_item("server_state",
"Values: UNSENT, IN_PROGRESS, OVER
- Initially UNSENT
- Set by scheduler to IN_PROGRESS when send result
- Set by scheduler to OVER when get reply from client
- Set by transitioner to OVER if now > result.report_deadline
- Set by transitioner to OVER if WU has error condition
and result.server_state=UNSENT
- Set by validator to OVER if WU has canonical result
and result.server_state=UNSENT
"
);
list_item("outcome",
"Values: SUCCESS, COULDNT_SEND, CLIENT_ERROR, NO_REPLY, DIDNT_NEED.
Defined iff result.server_state=OVER
- Set by scheduler to SUCCESS if get reply and no client error
- Set by scheduler to CLIENT_ERROR if get reply and client error
- Set by transitioner to NO_REPLY if server_state=IN_PROGRESS
and now Set by transitioner to DIDNT_NEED if WU has error condition
and result.server_state=UNSENT
- Set by validator to DIDNT_NEED if WU has canonical result
and result.server_state=UNSENT
"
);
list_item("client_state",
"Records the client state (upload, process, or download)
where an error occurred.
Defined if outcome is CLIENT_ERROR.
"
);
list_item("file_delete_state",
"
- Initially INIT
- Set by transitioner to READY if this is the canonical result,
and file_delete_state=INIT,
and wu.assimilate_state=DONE,
and all the results have server_state=OVER,
and all all the results with outcome=SUCCESS have validate_state<>INIT
- Set by transitioner to READY if wu.assimilate_state=DONE
and result.outcome=CLIENT_ERROR
or result.validate_state!=INIT
"
);
list_item("validate_state",
"
Defined iff result.outcome=SUCCESS
- Initially INIT
- Set by validator to VALID if outcome=SUCCESS and matches canonical result
- Set by validator to INVALID if outcome=SUCCESS and doesn't match canonical result
- Set by validator to ERROR if outcome=SUCCESS and
had a permanent error trying to read an output file,
or an output file had a syntax error.
- Set by validator to INCONCLUSIVE if check_set()
didn't find a consensus in a set of results containing this one.
"
);
list_end();
echo "
Result invariants:
- Eventually server_state = OVER.
- Output files are eventually deleted.
Notes on deletion of output files:
- Non-canonical results can be deleted as soon as the WU is assimilated.
- Canonical results can be deleted only when all results have server_state=OVER and all success results are validated.
- If a result reply arrives after its timeout,
the output files can be immediately deleted.
How do we delete output files that arrive REALLY late?
(e.g. uploaded after all results have timed out, and never reported)?
Possible answer:
let X = create time of oldest unassimilated WU.
Any output files created before X can be deleted.
";
page_tail();
?>