Self-Managed Block Storage Scheduling for OpenStack-based Cloud

Abstract

Abstract Software-defined storage (SDS) enables cloud providers to abstract the storage resources from underlying physical hardware by making storage resources programmable. It enhances the scalability and management of storage resources for better efficiency. Software-defined block storage component in OpenStack is referred to as cinder. The storage scheduling algorithm in OpenStack cinder is inefficient and one dimensional as it considers the capacity of storage disks as the only parameter to schedule the request. This compromises QoS to the customer. This paper presents the design of a self-managed block storage scheduling model. Proposed algorithms considers the performance attributes like read and write IOPS of storage hosts and classifies hosts by generating a quality status using machine learning classifiers. Further, this status is used in conjunction with a best -fit algorithm to make a final scheduling decision. Private cloud test-bed with 5 cinder hosts is set up to test scheduler performance. Results clearly show that the proposed model outperforms the default scheduler by evenly distributing the volume requests across the cinder nodes. The proposed model also maintains the QoS by providing the best available storage backend for every volume request.