Proximity-Aware Traffic Routing in Distributed Fog Computing Platforms

Abstract

Container orchestration engines such as Kubernetes do not take into account the geographical location of application replicas when deciding which replica should handle which request. This makes them ill-suited to act as a generalpurpose fog computing platforms where the proximity between end users and the replica serving them is essential. We present proxy-mity, a proximity-aware traffic routing system for distributed fog computing platforms. It seamlessly integrates in Kubernetes, and provides very simple control mechanisms to allow system administrators to address the necessary trade-off between reducing the user-to-replica latencies and balancing the load equally across replicas. proxy-mity is very lightweight and it can reduce average user-to-replica latencies by as much as 90% while allowing the system administrators to control the level of load imbalance in their system.

Voilà: Tail-Latency-Aware Fog Application Replicas Autoscaler

“…However, this approach may dispatch requests to far-away replicas even if there exists nearby ones. To ensure that requests are received by nearby pods, Voilà relies on Proxy-mity, which overrides the routing rules to favorize replicas reachable through low-latency routes [7].…”

“…Voilà integrates seamlessly with Kubernetes, the de-facto standard container orchestration framework in clusters and data centers [12]. Kubernetes is also a promising basis for designing future-generation fog computing platforms [4], [7], [14], [24], [27]. Voilà continuously monitors the request workload produced by all potential traffic sources in the system, and uses efficient algorithms to determine the number and location of replicas that are necessary to maintain the application's QoS within its expected bounds despite potentially large variations in the request workload characteristics.…”

Latency-aware Resource Management at the Edge

“…Ahvar [67] Aït-Salaht [85] Bittencourt [78] Borgia [86] Borylo [71] Boubin [87] Chen [81] De Maio [88] Elgazar [83] Fahs [89] Fricker [72] Gu [80] Habak [20] Liu [77] Liu [50] Mahmud [90] Penner [52] Rodrigues [73] Sardellitti [82] Singh [40] Skarlat [46] Sonmez [91] Tang [79] Tärneberg [75] Wang [45] Wang [70] Wang [92] Wang [93] Xia [94] Yi [76] Zamani [74] 3.4. Objective user.…”

“…for load distribution [72]. Some works consider dynamic resource provisioning [71], VM placement [80], or container placement [89].…”

Plataformas de Fog Computing: da Teoria à Prática

“…No paradigma fog computing, em contraste com a computação em nuvem tradicional, os recursos e serviços estão distribuídos em posições estratégicas em uma área geográfica potencialmente ampla, para atender a uma grande variedade de usuários [18]. Desta forma, segundo [25], as plataformas precisam escolher de forma adequada os nós fog que executarão as aplicações, levando em consideração não somente as posições geográficas, mas também os critérios de QoS (Qualidade de Serviço, do inglês Quality of Service) demandados pela aplicação. Tal escolha precisa ser transparente para as aplicações e usuários, i.e., os usuários não precisam saber da localização onde suas aplicações serão executadas.…”