Topic: asynchronous processing
Topic: discrete vs. continuous
Topic: events
Topic: local vs. global
Topic: Newtonian physics
Topic: Petri net transitions and events
Topic: reliable communication
Topic: special relativity
Topic: synchronized processing
Topic: temporal relationships
Topic: time
Topic: timestamps
Topic: topology
Topic: version control
Topic: version identification
| |
Summary
Every sequence of events, every activity, makes time, and every interaction takes time. Time is then a local relation between states and between events, with post-states of an interaction happening after its pre-states.
Event time is seen as a understanding of reality where phenomena, once they happen, can not unhappen. A world without events is a world without time; so is a world with only one event. But a world with many events, richly interrelated, is a world inseparable from time. (cbb 6/83)
Subtopic: time as abstraction
Quote: time does not exist in its own right; it is part of things at motion and at rest; what happens to things in the past, present, and future [»lucr_55]
| Quote: there is no absolute space, no absolute time, no simultaneity across distance, not even geometry; these concepts are no more than a language [»poinH_1902, OK]
| Quote: to say A happens before B is equivalent to a new thing U whose first tick is A and second tick is B [»cbb_1980, OK]
| Subtopic: time as simultaneous events
Quote: all judgments of time concern simultaneous events [»einsA9_1905]
| Quote: choice uncertainty principle -- can not chose between near-simultaneous events within a deadline [»dennPJ11_2007]
| Subtopic: time as responsibility
Quote: every time you take on anything, it requires action within a finite interval of time; like spinning plates in a circus act [»holtAW_1997]
| Subtopic: clocks
Quote: the propagation of light assigns equal tracts to equal time intervals [»einsA_1923]
| Quote: clocks at rest relative to each other are synchronized if, by definition, the time for light to travel from A to B is the time for it to travel from B to A [»einsA9_1905]
| Quote: if ideal clocks go at the same rate, they always go at the same rate; e.g., sharp spectral lines from atoms of the same chemical element; a foundation of 4-d space-time
| Quote: objective time requires a local time (clock) that is extended to events in the whole space [»einsA3_1936]
| Quote: without physical time, there is no way to distinguish a failed process from one that is just pausing between events [»lampL7_1978]
| Quote: use heartbeat messages for communication between master and chunkservers
| Subtopic: time as state transitions
Quote: programming is not interested in the state of the machine at every moment; only at prepulses
| Quote: can judge incidence between states or transitions but not the coincidence of either; e.g., a clock dial is between 1 and 2 vs. exactly 1; i.e., continuous motion [»petrCA_1996]
| Quote: time is a strictly, local relation between states [»petrCA1_1966]
| Quote: logical execution time (LET) separates the observable real-time behavior from its physical execution; execute between logical release and terminate events; outputs have the value of the previous execution [»farcE6_2005]
| Quote: a system's transitions and interactions impose a temporal ordering on events [»holtAW3_1979]
| Quote: a switching element only defines the temporal successor of a state [»petrCA1_1966]
| Quote: a hiker going between valleys is like a clock whose ticks are the hiker's transitions from valley to ridge [»holtAW11_1980]
| Quote: time is activity; the ticking of a clock is the simplest activity; or is time a single tick [»cbb_1980, OK]
| Subtopic: time is succession of local events
Quote: time in a process is only identifiable by a succession of events; two processes in a temporal relationship but no absolute time [»mantMJ2_1983]
| Quote: a temporal relationship defines a partial ordering; events are concurrent if they are unordered [»holtAW_1970]
| Quote: a temporal relationship defines a directed acyclic graph [»holtAW_1970]
| Quote: every actor has a local time axis defined by messages arrival from other actors
| Quote: time is a (single?) ordered sequence of event times [»nygaK10_1983, OK]
| Quote: the local time axis defines acquaintances (other actors) known at an event [»hewiC8_1977]
| Quote: in DELTA time is defined by a global variable; next version of DELTA will use a petri net definition of time [»handP_1981]
| Subtopic: time in petri nets
Quote: Petri assumed that metrics do not exist for space, time, or any physical magnitude [»petrCA1_1966]
| Quote: a Petri net treats time as a partial order on event occurrence; no inherent measure of time [»peteJL9_1977]
| Quote: Petri nets feature proper concurrency; i.e., the priority of causality over temporal order, and partial independence [»petrCA_1996]
| Subtopic: derived time
Quote: flows can follow slower clocks; e.g., 'x when c' is x's values when c is true; most flow operators use the same clock [»benvA1_2003]
| Quote: flow operators deactivated if an operand clock is false [»benvA1_2003]
| Quote: CATOCS message ordering delays unrelated messages; happens-before does not imply dependency
| Quote: user-defined time, e.g., the date when the promotion committee approved the promotion [»snodR9_1986]
| Subtopic: time local to each process
Quote: implement time by a timestamp on each message; receiver advances its process time to be later than the timestamp [»lampL7_1978]
| Quote: a queued model of communication allows local time for each process and avoids issues of simultaneity [»stroRE5_1985]
| Quote: in process paradigm there is no global state or global time due to concurrent activity [»stroR10_1986]
| Quote: a message system defines a partial order of events; send event is before its receive event [»lampL7_1978]
| Quote: when a receiver receives a message its local time is after the local time of the sender when sending the message [»stroRE5_1985]
| Quote: order all system events by the process times at which they occur; break ties with a process id [»lampL7_1978]
| Subtopic: time as a sequence of versions
Quote: use object versions to implement logical clocks; i.e., changes to state; slower than message clock ticks [»cherDR12_1993]
| Subtopic: time as a counter
Quote: time in occam represented by a counter; different parts of a program may use different timers; easier than global time [»mayD_1987]
| Quote: a clock is an unbounded number of ticks, i.e., CLOCK = (tick -> CLOCK) [»hoarCA_1985]
| Quote: a clock is a body or system that counts a series of events which can be regarded as equal
| Quote: signal the occurrence of an event by incrementing its event count [»reedDP2_1979]
|
Related Topics
Topic: asynchronous processing (30 items)
Topic: discrete vs. continuous (47 items)
Topic: events (44 items)
Topic: local vs. global (29 items)
Topic: Newtonian physics (79 items)
Topic: Petri net transitions and events (21 items)
Topic: reliable communication (29 items)
Topic: special relativity (73 items)
Topic: synchronized processing (35 items)
Topic: temporal relationships (40 items)
Topic: time (49 items)
Topic: timestamps (19 items)
Topic: topology (29 items)
Topic: version control (34 items)
Topic: version identification (12 items)
|