2022
DOI: 10.48550/arxiv.2205.01944
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Joint Compute-Caching-Communication Control for Online Data-Intensive Service Delivery

Abstract: Emerging Metaverse applications, designed to deliver highly interactive and immersive experiences that seamlessly blend physical reality and digital virtuality, are accelerating the need for distributed compute platforms with unprecedented storage, computation, and communication requirements. To this end, the integrated evolution of next-generation networks (e.g., 5G and beyond) and distributed cloud technologies (e.g., fog and mobile edge computing), have emerged as a promising paradigm to address the interac… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
14
0

Year Published

2022
2022
2023
2023

Publication Types

Select...
4
1

Relationship

4
1

Authors

Journals

citations
Cited by 6 publications
(14 citation statements)
references
References 33 publications
0
14
0
Order By: Relevance
“…More recently, a number of studies have addressed the SFC optimization problem in dynamic scenarios, where one needs to make joint packet processing and routing decisions in an online manner [17]- [21]. The works in [17], [18] employ a generalized cloud network flow model that allows joint control of processing and transmission flows.…”
Section: A Related Workmentioning
confidence: 99%
See 1 more Smart Citation
“…More recently, a number of studies have addressed the SFC optimization problem in dynamic scenarios, where one needs to make joint packet processing and routing decisions in an online manner [17]- [21]. The works in [17], [18] employ a generalized cloud network flow model that allows joint control of processing and transmission flows.…”
Section: A Related Workmentioning
confidence: 99%
“…The works in [17], [18] employ a generalized cloud network flow model that allows joint control of processing and transmission flows. The works in [19]- [21] show that the traffic control problem in distributed cloud networks (involving joint packet processing and routing decisions) can be reduced to a packet routing problem on a properly constructed layered graph that includes extra edges to characterize the processing operations (i.e., packets pushed through these edges are interpreted as being processed by a service function). By this transformation, many control policies designed for packet routing can be extended to address cloud network control problems (i.e., packet processing and routing), especially those aiming at maximizing network throughput with bounded average delay performance.…”
Section: A Related Workmentioning
confidence: 99%
“…The previous propositions can be explained as follows: by Proposition 1, in general, the admissible policy spaces A 1 A 2 and A 2 A 1 ; Proposition 2 suggests that they lead to the same capacity regions by presenting an explicit identical characterization (12), where (12c) is interpreted as the generalized flow conservation law when considering the packets' lifetime; Proposition 3 further shows that P 1 and P 2 share the same flow space for any (λ, γ), which is a crucial property for the considered problem, where the two metrics of interest, i.e., timely throughput (7) and resource cost (8), are both linear functions of the flow assignment.…”
Section: B Relationships Between P 1 and Pmentioning
confidence: 99%
“…Previous studies have shown that the cloud network control problem [7]- [12], involving packet routing and processing decisions over a distributed computing network, can be connected to the packet routing problem in traditional communication networks via a properly constructed cloudaugmented or layered-graph formulation [13]. For packet routing, many dynamic control policies have been developed aimed at maximizing network throughput, including the celebrated back-pressure (BP) algorithm [14] and its extension, the Lyapunov drift-plus-penalty (LDP) control approach [15] that, in addition, optimizes network resource cost (e.g., energy expenditure).…”
Section: Introductionmentioning
confidence: 99%
“…The proliferation of real-time stream-processing applications such as augmented reality, telepresence, and industrial automation [3]- [5], is pushing the evolution of networking and cloud technologies in order to meet their stringent low latency and compute-intensive requirements [6]. Traditional approaches treat network and cloud resources separately, with fairly centralized core clouds handling the processing of compute-intensive tasks, while the network takes care of routing data streams from sources to the cloud, and back to their destinations.…”
Section: Introductionmentioning
confidence: 99%