Multi-cloud computing systems face significant challenges in ensuring acceptable performance while adhering to tenant budget requirements. This paper proposes a tenant budget-aware (tenant-centric) data replication framework for Multi-Cloud Computing (TCDRM). The proposed strategy dynamically creates data replicas based on predefined thresholds for response time, economic budget of the tenant and data popularity. TCDRM employs a heuristic replica placement algorithm that leverages the diverse pricing structures of multiple cloud providers. The TCDRM strategy aims to maintain the required performance without exceeding the tenant's budget by taking advantage of the capabilities offered by multicloud environments. The middleware considered acts as an intermediary between tenants and multiple cloud providers, facilitating intelligent replica placement decisions. To achieve this, the proposed TCDRM strategy defines strict thresholds for tenant budget and response time. A performance evaluation is conducted to validate the effectiveness of the strategy. The results show that our approach effectively meets tenant performance objectives while respecting their economic constraints. Bandwidth consumption is reduced by up to 78% compared to non-replicated approaches, and average response time for complex queries is decreased by 51%, all while adhering to tenant budget limitations.

翻译：多云计算系统在确保可接受性能的同时满足租户预算要求方面面临重大挑战。本文提出了一种面向多云计算的租户预算感知（以租户为中心）数据复制框架（TCDRM）。该策略基于预定义的响应时间阈值、租户经济预算及数据流行度动态创建数据副本。TCDRM采用启发式副本放置算法，充分利用多个云服务提供商的差异化定价结构。通过利用多云环境提供的功能，TCDRM策略旨在维持所需性能且不超出租户预算。所设计的中间件作为租户与多云提供商之间的中介，促进智能副本放置决策。为此，提出的TCDRM策略为租户预算和响应时间定义了严格阈值。通过性能评估验证了策略的有效性。结果表明，我们的方法在尊重租户经济约束的同时，能有效满足其性能目标。与非复制方法相比，带宽消耗降低达78%，复杂查询的平均响应时间减少51%，且全程遵守租户预算限制。