Greedy Nominator Heuristic: Virtual function placement on fog resources

Abstract

Fog computing is an intermediate infrastructure between edge devices (e.g., Internet of Things) and cloud systems that is used to reduce latency in real‐time applications. An application can be composed of a collection of virtual functions, between which dependency constraints can be captured in a service function chain (SFC). Virtual functions within an SFC can be executed at different geo‐distributed locations. However, virtual functions are prone to failure and often do not complete within a deadline. This results in function reallocation to other nodes within the infrastructure; causing delays, potential data loss during function migration, and increased costs. We proposed Greedy Nominator Heuristic (GNH) to address these issues. GNH is based on redundant deployment and failure tracking of virtual functions. GNH places replicas of each function at multiple locations—taking account of expected completion time, failure risk, and cost. We make use of a MapReduce‐based mechanism, where Mappers find suitable locations in parallel, and a Reducer then ranks these locations. Our results show that GNH reduces latency by up to 68%, and is more cost effective than other approaches which rely on state‐of‐the‐art optimization algorithms to allocate replicas.

Adaptive Edge-Cloud Environments for Rural AI

“…We use the Greedy Nominator Heuristic (GNH) [3] to solve the placement problem. GNH relies on redundant virtual function deployment and failure tracking.…”

“…Based on [3], the optimisation problem can be formulated as minimisation of C, R and O (based on symbols in table II).…”

“…Failure of function deployment, on the other hand, indicates that the function execution overlapped with the MTTR period of the field-side unit's MTBF-clock. Figure 5 shows service functions that were submitted over three different MTBF-clock periods [3].…”

“…A random walk begins with the selection of a random number of steps to be taken. The number of steps is [1][2][3][4][5][6][7][8][9][10]. After deciding on the steps, each step can be north, south, east, or west.…”

“…In our previous work, we described the Greedy Nominator Heuristic (GNH) [3], a distributed optimization algorithm for supporting function placement within a cloud-edge environment. GNH relies on redundancy in service functions and supports failure tracking on the connected computational resources.…”