Tiered Backup Archival in Multi-Tenant Cloud Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multi-tenant cloud computing environments, managing backup schedules and resources efficiently across multiple tenants with varying SLAs and constraints is challenging, leading to potential service agreement violations and increased infrastructure costs.
Innovation Solution
Implementing a fairness algorithm within the scheduler to account for backup schedules, resource constraints, and tenant priorities, allowing for the efficient sharing of physical storage and leveraging deduplication while prioritizing recovery tasks, and utilizing archival storage for less frequently accessed data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If backup data is stored in expensive high-speed storage for all tenants, then data retrieval speed is improved, but infrastructure cost increases significantly
Solution Approach 1:
The patent segments backup data into different categories based on access patterns and SLA requirements. Hot data (frequently accessed) is stored in high-speed storage, while cold data (rarely accessed) is moved to archival storage. This segmentation allows the system to optimize retrieval speed for critical data while reducing overall infrastructure costs by not expending high-speed storage resources on all data uniformly.
Solution Approach 2:
The patent applies local quality by providing different storage quality levels to different data based on their specific needs. Tenants with strict RTO requirements receive high-speed storage for their critical workloads, while tenants with more flexible requirements can use archival storage. This localized optimization ensures that only the necessary portions of the system incur high costs, while other portions use more economical storage solutions.
2Quantity of substance
If archival storage is used for all backup data, then infrastructure cost is reduced, but data retrieval speed deteriorates
Solution Approach 1:
The patent implements preliminary action by proactively identifying and archiving backup data that is unlikely to be accessed soon based on age, access patterns, and tenant-defined policies. This prevents the need to keep all data in expensive high-speed storage from the beginning, thereby reducing infrastructure costs while ensuring that data which might need fast retrieval remains in high-speed storage.
Solution Approach 2:
The patent applies dynamics by making the storage allocation flexible and adaptive rather than static. The system continuously monitors data access patterns and automatically moves data between high-speed and archival storage based on current needs. This dynamic approach allows the system to optimize the balance between cost and retrieval speed in real-time, adjusting to changing tenant requirements and data characteristics.
3Productivity
If backup resources are allocated without fairness considerations, then productivity increases for some tenants, but service agreement violations occur for others
Solution Approach 1:
The patent implements feedback mechanisms that continuously monitor backup progress, resource utilization, and SLA compliance across all tenants. When the system detects that a tenant is at risk of violating their SLA, it automatically adjusts resource allocation to prioritize that tenant's backup operations. This feedback loop ensures that high productivity for one tenant does not come at the expense of another tenant's reliability requirements.
Solution Approach 2:
The patent applies parameter changes by dynamically adjusting backup scheduling parameters, resource allocation weights, and priority levels based on current system state and tenant SLA requirements. Instead of fixed allocation, the system modifies these parameters in real-time to balance overall productivity with individual tenant reliability needs, allowing flexible optimization of both throughput and compliance.
4Quantity of substance
If deduplication is aggressively applied across all tenant data, then storage efficiency improves, but processing time increases
Solution Approach 1:
The patent applies partial action by selectively applying deduplication only to appropriate data sets rather than universally. The system identifies data that is suitable for deduplication based on characteristics such as redundancy levels, access patterns, and tenant preferences, and applies deduplication only to those portions. This partial application achieves storage efficiency gains without incurring the full processing time cost across all data.
Solution Approach 2:
The patent implements periodic action by scheduling deduplication operations at specific intervals rather than continuously or on-demand. The system performs deduplication during off-peak hours or in periodic batches, spreading the processing load over time. This approach reduces the impact on processing time during critical operations while still achieving storage efficiency improvements through regular deduplication cycles.
Data Source
AI summary
A system and method for backing up workloads for multiple tenants of a cloud computing system are disclosed. A method of backing up workloads for multiple tenants of a computing system includes triggering an archival process according to an archival policy set by a tenant, and executing the archival process by reading backup data of the tenant stored in a backup storage device of the computer system and transmitting the backup data to an archival store designated in the archival policy, and then deleting or invalidating the backup data stored in the backup storage device.