Local File System Translating Random-Access to Append Operations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed file systems (DFS) in cloud computing face limitations in providing random-access-write capabilities, which are essential for certain applications, and this hinders the scalability of storage resources to meet changing computing workloads.
Innovation Solution
Implementing a local file system on computing nodes that translates random-access read and write operation calls into append-based operations compatible with an append-only DFS, allowing data to be stored across storage nodes in a cloud network, thereby decoupling hosted storage from computational nodes and enabling dynamic scaling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a distributed file system is implemented as an append-only file system, then reliability and simplicity are improved, but random-access-write capability is lost
Solution Approach 1:
The patent introduces a translation layer (file system translation module) that acts as an intermediary between the POSIX file system interface and the append-only DFS. This translator converts random-access write operations into append operations, allowing the DFS to maintain its append-only nature while providing random-access capabilities to applications through translation.
Solution Approach 2:
The patent creates a virtual copy of the file system interface through translation. Instead of modifying the actual DFS storage structure, it creates a virtual layer that copies and translates operations, allowing random-access writes to be simulated as append operations in the underlying append-only system.
2Productivity
If dedicated storage is implemented for individual nodes, then random-access-write performance is improved, but scalability is hindered
Solution Approach 1:
The patent makes the storage system universal by allowing any computing node to provide storage services through the translation layer. Instead of dedicated storage per node, any node can host the DFS and serve random-access write needs for multiple applications, enabling storage resources to be dynamically allocated and scaled across the cluster.
Solution Approach 2:
The patent introduces dynamic scalability through the translation layer that can be deployed on any computing node. Storage resources can be dynamically allocated to different nodes based on workload demands, allowing the system to adapt to changing computing tasks without static storage dedications.
3Adaptability or versatility
If a translation layer is added to enable random-access operations on DFS, then adaptability is improved, but device complexity increases
Solution Approach 1:
The patent extracts the complexity of random-access operation handling from the core DFS storage system and places it in a separate translation layer. This separates the simple append-only DFS logic from the complex translation logic, allowing the DFS to remain simple while the translation layer handles the complexity of converting random-access operations.
Data Source
AI summary
Methods and systems implement local file systems of computing nodes which translate random-access read and random-access write operation calls to append-based operation calls in accordance with a distributed file system (“DFS”) implemented across storage nodes of a cloud network. Computer-executable applications running on the computing nodes may generate kernel-level read and write system calls by application programming interfaces (“APIs”) such as the Portable Operating System Interface (“POSIX”) standard, to a local file system. The local file system may translate these read and write system calls to file operations at a DFS implementing an append-only file system, as well as perform storage reclamation upon the DFS periodically and/or upon storage thresholds being exceeded. Storage for computing workloads on the computing nodes may be decoupled from individual computing nodes, in comparison to conventional architectures for computing clusters and storage clusters in cloud networks, enabling storage to be scaled without static allocations.