Scheduling of early quantum tasks

Abstract

An Early Quantum Task (EQT) is a Quantum EDF task that has shrunk its first period into one quantum time slot. Its purpose is to be executed as soon as possible, without causing deadline overflow of other tasks. We will derive the conditions under which an EQT can be admitted and can have an immediate start. The advantage of scheduling EQTs is shown by its use in a buffered multimedia server. The EQT is associated with a multimedia stream and it will use its first invocation to fill the buffer, such that a client can start receiving data immediately 1 .

Performance analysis of a dependable scheduling strategy based on a fault-tolerant grid model

“…Some of existing scheduling strategies are based on the static pattern, such as Rational Rate-Monotonic (RRM) [7], Early Quantum Task (EQT) [8]and Earliest Deadline First (EDF) [9]. Usually they only consider the difference in solving task, while the difference of computing node is rarely considered.…”

Eligible earliest deadline first: Server-based scheduling for master-slave industrial wireless networks

“…Since T is also the effective worst-case duration for a transaction, each transaction duration takes exactly one slot or less (i.e. Ci = 1), leading to a so-called 'Quantum' scheduling instance [26]. With this in mind we may model each transaction simply by its period Ti, and the slave ID Si.…”

“…For this, we use the EDF algorithm [19] [25], in which the ready job with the nearest absolute deadline is always allocated the current slot. In the case of Quantum EDF scheduling, this algorithm is optimal and guarantees that all deadlines will be respected (in the error-free case) if and only if the condition U  1 holds [26]. Suppose that the primary transaction is processed during some specific slot.…”

“…In the current context, the spare capacity of the overall wireless network is the utilization which is unused by the primary transactions in the error-free case, having the value (1-U). Although it is difficult to implement an optimal server in the general case [25], when Quantum scheduling is employed as is the current case the optimal server has a very simple form [26]. First, define the replenishment period of the server as the integer Ts, as follows:…”

“…If the time at which the last aperiodic transaction was attempted occurred at time tl, the next pending aperiodic transaction may be immediately executed at any time slot equal to or subsequent to the time tn = tl + Ts, without impacting upon the timing properties of any primary transactions [26]. This server, with period selected according to (6), is related to the improved total bandwidth server for general EDF task scheduling (e.g.…”