Xiao Liu PhD Thesis.pdf - Faculty of Information and Communication ...

More documents

Recommendations

Info

(deadlines) and local temporal constraints for workflow segments and/or individual activities (milestones) are required to be set as temporal QoS constraints in cloud workflow specifications. The problem for the conventional QoS constraint setting strategies lies in three aspects: first, estimated workflow activity durations (based on user experiences or simple statistics) are often inaccurate in cloud workflow system environments; second, these constrains are not well balanced between user requirements and system performance; third, the time overheads of the setting process is non-trivial, especially for large numbers of local temporal constraints. To address the above issues, a statistical time-series pattern based forecasting strategy is first investigated to predict the duration intervals of cloud workflow activities. Afterwards, based on the weighted joint normal distribution of workflow activity durations, a probabilistic setting strategy is applied to assign coarse-grained temporal constraints through a negotiation process between service users and service providers, and then the fine-grained temporal constraints can be propagated along scientific cloud workflows in an automatic fashion, i.e. with very small time overheads. At scientific cloud workflow runtime, in Component 2, the state of cloud workflow execution towards specific temporal constraints, i.e. temporal consistency, is monitored constantly by the checkpoint selection and temporal verification component. The two major limitations of conventional checkpoint selection and temporal verification are as follows: first, the selection of multiple types of checkpoints and the verification of multiple types of temporal consistency states incur huge cost but most of them are actually unnecessary; second, though the violations of multiple temporal consistency states can be verified, there is no clear indication for the level of temporal violations, i.e. it does not support the quantitative measurement of temporal violations. To address the above issues, with our probability based temporal consistency model, the probability range for statistically recoverable and non-recoverable temporal violations are defined in the first place. Afterwards, we monitor cloud workflow executions through the following two steps: first, a minimum probability time redundancy based temporal checkpoint selection strategy determines the activity points where potential temporal violations take place; second, probability temporal consistency based temporal verification is conducted on a checkpoint to check the current temporal consistency state and the types of 8
temporal violations, viz. recoverable or non-recoverable. Finally, in Component 3, detected temporal violations are handled. The conventional temporal verification philosophy believes that temporal violation handling should be triggered on every checkpoint, i.e. whenever a temporal violation is detected. However, the number of checkpoints for large scale scientific workflow applications is often huge in dynamic cloud computing environments, and thus results in significant cost on the handling of temporal violations. Therefore, we design a cost-effective adaptive temporal violation handling point selection strategy to further decide whether a checkpoint should be selected as a temporal violation handling point, i.e. temporal violation handling is necessary. Afterwards, when temporal violation handling points are selected, some temporal violation handling strategies are triggered to tackle the detected temporal violations. So far, temporal violation handling strategies in scientific cloud workflow systems have not been well investigated and the existing strategies are either of low performance or high cost. Therefore, in our temporal framework, an innovative temporal violation handling strategy is defined which consists of three levels of temporal violations (from minor to major) and their corresponding violation handling strategies (from weak capability to strong capability). In such a case, different levels of temporal violations can be handled according to the capability of different temporal violation handling strategies with appropriate costs. Among many others, we focus on costeffective light-weight metaheuristics based workflow rescheduling strategies for statistically recoverable temporal violations. Specifically, with the design of a general two-stage local workflow rescheduling strategy, two representative metaheuristic algorithms including Genetic Algorithm (GA) and Ant Colony Optimisation (ACO) are adapted and implemented as candidate temporal violation handling strategies. 1.4 Overview of this thesis In particular, this thesis deals with the design of a comprehensive temporal framework which includes a set of new concepts, strategies and algorithms for the support of high QoS over the whole lifecycle of scientific cloud workflow applications. The thesis structure is depicted in Figure 1.2. 9
Page 1 and 2: A Novel Probabilistic Temporal Fram
Page 3 and 4: Declaration This thesis contains no
Page 5 and 6: Abstract Cloud computing is a lates
Page 7 and 8: Component 2, the state of scientifi
Page 9 and 10: http://dx.doi.org/10.1016/j.jpdc.20
Page 11 and 12: 17. X. Liu, J. Chen, and Y. Yang, A
Page 13 and 14: 4.2 RELATED WORK AND PROBLEM ANALYS
Page 15 and 16: STRATEGY ..........................
Page 17 and 18: FIGURE 7.4 EXPERIMENTAL RESULTS (NO
Page 19 and 20: Chapter 1 Introduction This thesis
Page 21 and 22: Clearly, if the execution time of t
Page 23 and 24: to seek all the candidates. Further
Page 25: e unnecessary. Hence, an effective
Page 29 and 30: In Chapter 2, we introduce the rela
Page 31 and 32: a two-stage searching process with
Page 33 and 34: QoS requirements. Generally speakin
Page 35 and 36: handling strategies employed in a s
Page 37 and 38: activity points to conduct temporal
Page 39 and 40: compensation, is believed as a suit
Page 41 and 42: successful completion. Specifically
Page 43 and 44: Chapter 4 and Chapter 5 respectivel
Page 45 and 46: etween time deficit compensation an
Page 47 and 48: service user’s requirements and t
Page 49 and 50: conducted for three times, i.e. sta
Page 51 and 52: PTDA+ACOWR as an example to introdu
Page 53 and 54: currently running. It is built on t
Page 55 and 56: workflow instance to determine whet
Page 57 and 58: framework for cost-effective delive
Page 59 and 60: pattern based forecasting strategy.
Page 61 and 62: 4.2 Related Work and Problem Analys
Page 63 and 64: 4.2.2 Problem Analysis In cloud wor
Page 65 and 66: significantly deteriorate the overa
Page 67 and 68: activity durations: characteristics
Page 69 and 70: specified with different means and
Page 71 and 72: Table 4.1 Notations Used in K-MaxSD
Page 73 and 74: 1) Duration series building As ment
Page 75 and 76: Table 4.5 Algorithm 4: Pattern Matc
Page 77 and 78:
to search for those activities whic
Page 79 and 80:
cycle. Here, we also trace back the
Page 81 and 82:
ecognition. Sliding Window ranks in
Page 83 and 84:
predicted value will be the one pre
Page 85 and 86:
segmentation algorithm is capable o
Page 87 and 88:
distributed soft real-time system,
Page 89 and 90:
Table 5.1 Overview on the Support o
Page 91 and 92:
2 2 can be denoted as N ( µ , σ )
Page 93 and 94:
mean iteration times or with some p
Page 95 and 96:
Figure 5.4 Choice Building Block No
Page 97 and 98:
activities may be omitted for the e
Page 99 and 100:
5.3.1 Calculating Weighted Joint Di
Page 101 and 102:
constraints until the constraint is
Page 103 and 104:
Therefore, it can be expressed as n
Page 105 and 106:
Table 5.3 Specification of the Work
Page 107 and 108:
and the constraint for activity X 1
Page 109 and 110:
ease of discussion and to avoid the
Page 111 and 112:
Unfortunately, conventional discret
Page 113 and 114:
The probability consistency range w
Page 115 and 116:
6.2.3 Temporal Checkpoint Selection
Page 117 and 118:
100 times each. All the activity du
Page 119 and 120:
normal distribution and correlated
Page 121 and 122:
Figure 6.3 Checkpoint Selection wit
Page 123 and 124:
work and problem analysis. Section
Page 125 and 126:
expected time redundancy of subsequ
Page 127 and 128:
the probability for self-recovery i
Page 129 and 130:
skipped if self-recovery applies. A
Page 131 and 132:
To evaluate the average performance
Page 133 and 134:
activities and around 300 violation
Page 135 and 136:
Figure 7.3 Experimental Results (No
Page 137 and 138:
Figure 7.5 Cost Reduction Rate vs.
Page 139 and 140:
8.1 Related Work and Problem Analys
Page 141 and 142:
scientific processes, human interve
Page 143 and 144:
alone does not involve any time def
Page 145 and 146:
swap slower machines in the active
Page 147 and 148:
TD( a p ) L{ ( ai , R j ) | i = p +
Page 149 and 150:
For example, in Figure 8.2, local w
Page 151 and 152:
value to sample all of the solution
Page 153 and 154:
mapping task a i to resource R j fr
Page 155 and 156:
have done some simulation experimen
Page 157 and 158:
violations can still be recovered b
Page 159 and 160:
PTDA+ACOWR for Level III Violations
Page 161 and 162:
an integer from 1 to 5 where the ex
Page 163 and 164:
D: Definition for Fitness Value Fit
Page 165 and 166:
strategy; and End(Rescheduling) is
Page 167 and 168:
(a) Optimisation Ratio on Total Cos
Page 169 and 170:
makespan. In our two stage workflow
Page 171 and 172:
espectively. It is clearly that ACO
Page 173 and 174:
comparison results on the violation
Page 175 and 176:
percentiles. The average global vio
Page 177 and 178:
equivalent times of ACOWR are calcu
Page 179 and 180:
Chapter 9 Conclusions and Future Wo
Page 181 and 182:
the real world, simulation experime
Page 183 and 184:
ased temporal consistency model. Ac
Page 185 and 186:
the whole lifecycle support for hig
Page 187 and 188:
which can be further investigated i
Page 189 and 190:
Bibliography [1] W. M. P. van der A
Page 191 and 192:
Systems", ACM Trans. on Software En
Page 193 and 194:
conjunction with 23 rd Parallel and
Page 195 and 196:
Elsevier, vol. 84, no. 3, pp. 354-3
Page 197 and 198:
Distributed Computing Systems", Jou
Page 199 and 200:
Appendix: Notation Index Symbols α
Page 201 and 202:
PTR ( U ( SW ), ( a p + , a p + 1 m
show all

Xiao Liu PhD Thesis.pdf - Faculty of Information and Communication ...

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?