| 50 | |
| 51 | |
| 52 | {{{ |
| 53 | <report_grace_period>x</report_grace_period> |
| 54 | }}} |
| 55 | A "grace period" for task reporting. A task is considered time-out (and a new replica generated) if it is not reported by client_deadline + x. |
| 56 | |
| 57 | {{{ |
| 58 | <user_filter>0|1</user_filter> |
| 59 | }}} |
| 60 | If set, use the "batch" field of workunits to select which user is allowed to process the job. If batch is nonzero, only send the job to the user with that ID. |
| 61 | |
| 62 | {{{ |
| 63 | <workload_sim>0|1</workload_sim> |
| 64 | }}} |
| 65 | Use a more expensive, but more accurate, method to decide whether hosts can complete jobs within their delay bound. |
| 66 | |
| 67 | === App version selection === |
| 68 | |
56 | | <report_grace_period>x</report_grace_period> |
57 | | }}} |
58 | | A "grace period" for task reporting. A task is considered time-out (and a new replica generated) if it is not reported by client_deadline + x. |
59 | | |
60 | | {{{ |
61 | | <user_filter>0|1</user_filter> |
62 | | }}} |
63 | | If set, use the "batch" field of workunits to select which user is allowed to process the job. If batch is nonzero, only send the job to the user with that ID. |
64 | | |
65 | | {{{ |
66 | | <workload_sim>0|1</workload_sim> |
67 | | }}} |
68 | | Use a more expensive, but more accurate, method to decide whether hosts can complete jobs within their delay bound. |
| 75 | <version_select_random_factor>X</version_select_random_factor> |
| 76 | }}} |
| 77 | In predicting which app version will be faster for a given host, |
| 78 | multiple the projected FLOPS by a uniform random variable |
| 79 | with mean 1 and this standard deviation (default 0.1). |