LJWOA: A Lévy-Flight-Enhanced Jaya–Whale Hybrid for Latency- and Power-Aware Microservice Placement in Fog-Cloud Environments

Placing containerized microservices across a fog-cloud continuum requires jointly optimizing network latency, node power consumption, and load balance. This paper studies LJWOA, a three-stage hybrid metaheuristic that adds a Jayastyle elitist refinement and a Lévy-flight restart of the weakest individuals to the Whale Optimization Algorithm (WOA). Evaluation uses six benchmark functions (dimension 30) and synthetic fog-cloud placement instances with 50–300 microservices (CPU workload 200–1500 MI, payload 1–50 MB) on 12–36 heterogeneous fog and cloud nodes, with 15–30 independent runs per setting and Holm-corrected Wilcoxon signed-rank tests. In the original experiments LJWOA attains a lower weighted objective than Jaya, GWO, and PSO at all five scales and than WOA at 50 and 100 microservices, is statistically indistinguishable from WOA at 150–200, and is significantly worse at 300. Additional experiments requested during review use an equal function-evaluation budget and an independent re-implementation: LJWOA remains ahead of Jaya, GWO, SMA, and NSGA-II at every scale and ties with PSO, but its advantage over WOA is not significant at 50–200 and reverses at 300 (0.612 vs. 0.576), and GA and L-SHADE reach lower objective values at all scales. The Lévy stage improves on the twostage variant on held-out instances (raw p < 0.01 at every scale), whereas the stage order has no consistent effect. LJWOA costs 1.36–1.43× WOA per iteration, and all algorithms overload on average 53–91% of nodes under the soft capacity constraint. LJWOA is therefore a modest improvement within the WOA family rather than a state-of-the-art placement method.

Authors

Institutions

Publication Details

Journal
Global Journal of Computer Science and Technology
Published
2026-11-21
DOI
https://doi.org/10.34257/gjcstb278922
Primary Topic
Cloud Computing and Resource Management
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

LJWOA: A Lévy-Flight-Enhanced Jaya–Whale Hybrid for Latency- and Power-Aware Microservice Placement in Fog-Cloud Environments

Rashmi Keshri, Rohan Kukreti, Prince Kumar Raj, Rashmi Keshri Keshri
Global Journal of Computer Science and Technology
Cloud Computing and Resource Management
article

LJWOA: A Lévy-Flight-Enhanced Jaya–Whale Hybrid for Latency- and Power-Aware Microservice Placement in Fog-Cloud Environments

Rashmi Keshri, Rohan Kukreti, Prince Kumar Raj, Rashmi Keshri Keshri
article en

Abstract

Placing containerized microservices across a fog-cloud continuum requires jointly optimizing network latency, node power consumption, and load balance. This paper studies LJWOA, a three-stage hybrid metaheuristic that adds a Jayastyle elitist refinement and a Lévy-flight restart of the weakest individuals to the Whale Optimization Algorithm (WOA). Evaluation uses six benchmark functions (dimension 30) and synthetic fog-cloud placement instances with 50–300 microservices (CPU workload 200–1500 MI, payload 1–50 MB) on 12–36 heterogeneous fog and cloud nodes, with 15–30 independent runs per setting and Holm-corrected Wilcoxon signed-rank tests. In the original experiments LJWOA attains a lower weighted objective than Jaya, GWO, and PSO at all five scales and than WOA at 50 and 100 microservices, is statistically indistinguishable from WOA at 150–200, and is significantly worse at 300. Additional experiments requested during review use an equal function-evaluation budget and an independent re-implementation: LJWOA remains ahead of Jaya, GWO, SMA, and NSGA-II at every scale and ties with PSO, but its advantage over WOA is not significant at 50–200 and reverses at 300 (0.612 vs. 0.576), and GA and L-SHADE reach lower objective values at all scales. The Lévy stage improves on the twostage variant on held-out instances (raw p < 0.01 at every scale), whereas the stage order has no consistent effect. LJWOA costs 1.36–1.43× WOA per iteration, and all algorithms overload on average 53–91% of nodes under the soft capacity constraint. LJWOA is therefore a modest improvement within the WOA family rather than a state-of-the-art placement method.

Global Journal of Computer Science and Technology
Uttaranchal University (IN)
Cloud Computing and Resource Management
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon
LJWOA: A Lévy-Flight-Enhanced Jaya–Whale Hybrid for Latency- and Power-Aware Microservice Placement in Fog-Cloud Environments — Rashmi Keshri, Rohan Kukreti, et al. · Global Journal of Computer Science and Technology (2026) | TGRS Research Map | TGRS