Hardware Reference
In-Depth Information
Semaphore
A kernel data structure used to synchronize the execution of concurrent
jobs.
Server
A kernel process dedicated to the management of a shared resource.
Service time
The amount of time required by the processor to execute a job without
interruption. It is also called
computation time
or
processing time
.
Shared resource
A resource that is accessible by two or more processes.
Slack time
The maximum delay that a job can experience after its activation and still
complete within its deadline. At the arrival time, the slack is equal to the relative
deadline minus the computation time (
D
−
C
). It is also called
laxity
.
Soft task
A task whose instances should be possibly completed within their dead-
lines, but no serious consequences occur if a deadline is missed.
Sporadic task
An aperiodic task characterized by a minimum interarrival time be-
tween consecutive instances.
Start time
The time at which a job starts executing for the first time.
Starvation
A phenomenon for which an active job waits for the processor for an
unbounded amount of time.
Static scheduling
A method in which all scheduling decisions are precomputed off-
line, and jobs are executed in a predetermined fashion, according to a time-driven
approach.
Synchronization
Any constraint that imposes an order to the operations carried out
by two or more concurrent jobs. A synchronization is typically imposed for satis-
fying precedence or resource constraints.
Tardiness
The interval of time in which a job stays active after its deadline. It is
also called
exceeding time
.
Task
A computation in which the operations are executed by the processor one at a
time. A task may consist of a sequence of identical jobs, also called instances. The
word
process
is often used as a synonym.
Search WWH ::
Custom Search