Browse Prior Art Database

Method and System for Restoring Scalable Files in a Cluster File System

IP.com Disclosure Number: IPCOM000243123D
Publication Date: 2015-Sep-16
Document File: 5 page(s) / 43K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method and system is disclosed for scalably restoring files and directories in a cluster file system.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 31% of the total text.

Page 01 of 5

Method and System for Restoring Scalable Files in a Cluster File System


Disclosed is a method and system for scalably restoring files and directories in a cluster file system. The system includes one or multiple compute-storage nodes coupled via network and the system provides a shared file system. Each compute-storage node hosts a backup client whereby backup clients are connected to a backup server via a network. The file system provides functions to create and store files. The backup client is configured to send data to a backup sever to create copies of existing files and directories with the purpose of data protection. Typically, the backup client is configured to send the data to the backup server in highly efficient multiple parallel data streams. Additionally, the backup client is also configured to restore data from the backup server when required. Generally, the backup client is also configured to restore the data from the backup server in a single data stream.

In an implementation of the method and system, an Efficient Scalable Restore (ESR) system is provided which includes one or multiple compute-storage nodes coupled via network, and a shared file system. Each compute-storage node hosts a restore client whereby the restore clients are connected to a backup server via a network. The restore client is configured to restore data efficiently from the backup server in terms of the order of the data restored and in terms of the number of bytes to be restored. The restore client is configured to restore the data from the backup server in multiple data streams in parallel to maximize the effective data restore rate. The ESR can be configured in various ways.

The ESR system is configured to interact with a user to generate a list of files to restore. The user provides a selection statement for files (wild carded list or selected list) to be restored. The ESR system queries a local database that holds all recent backup data information and lists all files matching the select statement. Thereafter, the ESR system presents the generated list to the user. The user agrees to the list or requests a new list based on incompleteness. If the user requests a new list then the ESR system queries the backup server database that holds all historical backup data and completes the list for the user. The user can also shorten the list if extraneous restore candidates are added from the query of all historical backup data. When the user agrees to the current list, the list is stored for further processing.

Once the restore candidate list has been approved by the user the system of ESR is configured to connect to the backup server and query for required information for the files and directories stored in a candidate file list. The information is as follows:

1


Page 02 of 5


1. Media Class (MC) of the storage media used to store the file or directory on the backup server
2. Volume Identifier (VID) of the volume (e.g. tape cartridge) used to store...