In a cloud environment, a cloud broker plays a vital role as an intermediary between cloud customers and providers, resolving issues and facilitating negotiations to balance customer preferences and provider profits. Over the past few years, numerous research articles have either directly or indirectly examined this area. Conducting a Systematic Mapping Study (SMS) on cloud brokerage is highly motivating as it offers a high-level overview of the research landscape, identifying trends, topics, gaps, and patterns within this dynamic and crucial field. This article presents an SMS conducted to categorize existing research, highlight underexplored areas, and map out the evolution and current state of cloud brokerage, providing valuable insights for researchers and practitioners. The SMS identified 91 relevant and reputable search spaces (journals and conferences) and 634 high-quality articles published from 2009 to 2022. Furthermore, we formulated and addressed eight significant research questions to clarify various aspects of the cloud broker field. The extracted information from the selected articles is included in a supplementary file, available online, offering valuable insights for research teams and developers interested in this domain.

Scheduling workflows in cloud environments is an important issue that many types of research have been conducted in this field. However, these approaches often focus on single workflow scheduling while the need for scheduling multiple workflows is growing. This study aims at presenting a cloud broker for executing Deadline-constrained Periodic scientific Workflows (BDPW). BDPW acts as a Workflow as a Service (WaaS) broker and uses both reserved and on-demand resources in order to minimize the monetary cost of renting resources from a cloud provider. Furthermore, BDPW uses container technology by executing multiple containerized tasks on the same Virtual Machine (VM) to decrease the provisioning delay of VMs. The proposed broker uses a hybrid scheduling method, i.e., static planning and dynamic scheduling. The static planner uses resource leveling problem (RLP) to provide a scheduling plan and also recognizes the number of reserved resources that should be leased from a provider. Then, the dynamic scheduler tries to assign tasks to the reserved resources based on the primary static plan and leases on-demand instances if necessary. Also, it may make changes to the primary plan due to uncertainties in the task runtimes. The experimental results in CloudSim show that BDPW outperforms baseline algorithms in terms of monetary cost.