Skip to main content

Advances, Systems and Applications

Table 1 The main parameters table

From: Reliability-aware failure recovery for cloud computing based automatic train supervision systems in urban rail transit using deep reinforcement learning

Parameter

Definition

M/\(\mathcal {G}^M\)/m

Number/Node set/Physical node

L/ \(\mathcal {G}^L\) /\(l_{mn}\)

Number of physical link/Physical link set/Physical link

H/ \(\mathcal {H}\) /h

Number of Resources/Resource type set/Resource type

\(A_m^h\) / \(A_{mn}^{BW}\)

Maximum customizable resources h for node m / Maximum customizable bandwidth for link mn

\(Q_m^h(t)\) / \(Q_{mn}^{BW}(t)\)

Utilization ratio of resource h in node m / bandwidth in physical link mn at time slot t

t/\(\sigma\)

Index/duration of time slot

S/\(\mathcal {S}\)/s

Traditional ATS service on cloud computing

\(\mathcal {O}_s\)/\(O_s\)

The set/number of sequenced microservice for ATS service s

\(V_o^s\)

The o-th microservice in service s

\(\mu _s\)/\(\vartheta _s\)

Maximum tolerable interruption time / traffic rate of traditional ATS service s

\(\varphi _{(o,s)}^h\) / \(M_o^s\)

The number of resource type h needed / cost of a backup microservice \(V_o^s\)

\(\chi _o^s(t)\)

The influence ratio of the backup placement to \(V_o^s\)

\(X_o^s(t)/b_o^s(t)/D_o^s(t)\)

The packet size/bandwidth/delay of synchronization link of \(V_o^s\)

\(k_{(o,s)}^m(t) / \check{k}_{(o,s)}^m(t)\)

The backup placement of \(V_o^s\) / The placement of \(V_o^s\)

\(p_o^s(t)\)

If \(V_o^s\) is supported by backup

\(\alpha _o^s(t)\)

The recovery procedure on \(V_o^s\)