A Fuzzy Logic Approach for Resources Allocation in Cloud Data Center

Abstract

The rapidly increasing demand of Cloud services, asking for a flexible and dynamic design of the Cloud, has become a major challenge in DC deployment. Classical Traffic Engineering approaches are no longer enough to deal with the efficient use of IT and network resources in this highly distributed scenario. To cope with this issue, we propose two Fuzzy logic controllers for DC resource allocation based on Mamdani and Sugeno inference processes, that are able to take advantage of simple heuristic rules for efficient virtual machines allocation. To test the effectiveness of the proposed controllers we compare their performance with two variants of Multi-objective allocators as well as a simple Mono-dimensional algorithm. Preliminary simulation tests validate our proposal in terms of number of allocated requests and average server resource utilization.

Proactive, uncertainty-driven queries management at the edge

“…The integration of a query processing subsystem into a distributed database management system enhanced by FL is used for analyzing query response time across fragmentations of global relations [54]. Another research effort, presented in [4], proposes the adoption of FL controllers for resource allocation taking advantage of simple heuristic rules for efficient virtual machines allocation in a distributed setup. FL controllers adopt a knowledge base having the form of a set of fuzzy rules that depict the uncertainty management processing of inputs.…”

“…Comparing our model with other schemes in the respective literature, we can note the following. Our model is based on interval Type-2 FL sets while other efforts adopt Type-1 sets [4], [62], thus, they provide limited uncertainty management. Additional efforts deal with the incorporation of the FL upon databases [47].…”

“…We adopt these data to feed the load parameter l. The remaining parameters take values as described for Datasets 1 & 2; (iv) Dataset 4. The trace reported by [65] also retrieved by the UCI Machine Learning Repository 4 The dataset is created upon the energy analysis of twelve (12) different building shapes. Buildings differ w.r.t.…”

Type Ii Fuzzy Logic Controllers for VM Management and Task Assignment in Cloud System

“…In [21] the fuzzy logic controller based VM selection algorithm to assign the upcoming task, under imprecise memory, bandwidth and disk space specifications. A simple VM allocation method using Fuzzy logic controllers based on Mamdani and Sugeno inference processes is presented in [22]. Type-I and type-II fuzzy logic controllers for load balance and availability (virtual execution unit (VEU)) prediction are presented in [23] for virtual data centers (VDC) with the uncertain workload and uncertain availability of VEU nodes.…”

Power Consumption-Aware Virtual Machine Placement in Cloud Data Center

“…The former maps input values into one or more fuzzy sets, the latter produces a single conventional value that best represents the inferred fuzzy values. More specifically, our FLC includes the following modules (see [24] for more details):…”