Disk Storage Allocation for Virtual Machines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional Hadoop deployments face limitations in providing quality of service guarantees across tenants due to lack of enforcement of resource constraints, leading to over-commitment of resources and low resource utilization, especially in multi-tenant scenarios where multiple jobs are executed concurrently.
Innovation Solution
The system allocates outer blocks of rotational disk storage to virtual machines associated with I/O intensive workloads, using logical block addressing to optimize storage allocation and enhance I/O throughput in a virtual server cluster environment, thereby ensuring efficient resource utilization and stronger security and resource isolation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If resources are over-committed to multiple tenants, then resource utilization appears high, but quality of service guarantees deteriorate and I/O throughput decreases
Solution Approach 1:
The patent applies local quality by allocating different disk blocks based on their radial position - outer blocks are preferentially allocated to I/O intensive workloads while inner blocks serve other purposes. This creates localized optimization where high-performance storage regions are dedicated to workloads that benefit most from them, ensuring quality of service guarantees while maintaining high overall resource utilization through differentiated allocation strategies.
2Productivity
If outer blocks are allocated to I/O intensive workloads, then I/O throughput is enhanced, but storage allocation complexity increases
Solution Approach 1:
The system performs preliminary action by pre-identifying and pre-allocating outer blocks to I/O intensive workloads before actual I/O operations commence. The block allocation mechanism proactively assigns high-performance storage regions based on workload characteristics, eliminating the need for complex real-time dynamic allocation during I/O operations and simplifying the overall storage management complexity.
3Adaptability or versatility
If resource constraints are not enforced, then multi-tenant flexibility is maintained, but resource isolation and security deteriorate
Solution Approach 1:
The patent applies segmentation by dividing the storage resource pool into distinct segments - outer blocks are segmented and allocated to I/O intensive workloads while inner blocks remain available for other purposes. This segmentation enforces resource isolation between different tenant workloads, providing security guarantees while maintaining multi-tenant flexibility through differentiated access to appropriate storage segments based on workload requirements.
Data Source
AI summary
A method for allocating storage for a virtual machine includes receiving a request to allocate a block of disk storage on a physical disk to a virtual disk associated with the virtual machine. The method also includes identifying a plurality of available blocks on the physical disk, each of the plurality of available blocks defining a radial distance from a center axis of a platter of the physical disk. The method further includes determining which of the plurality of available blocks has a greatest radial distance from the center axis of the platter and, based on the determining, selecting an allocation block from the plurality of available blocks, the allocation block having the greatest radial distance from the center axis of the platter. The method also includes allocating the allocation block to the virtual machine.