Browse Prior Art Database

A disaster recovery solution combining with block and object storage

IP.com Disclosure Number: IPCOM000246845D
Publication Date: 2016-Jul-07
Document File: 5 page(s) / 132K

Publishing Venue

The IP.com Prior Art Database

Abstract

This invention proposes a method to use the object storage as the secondary site in the DR (Disaster Recovery) solution. With this method, there is NO extra change in the host. From the host's perspective, both the primary and secondary storage is block storage. And the host can access both the primary and secondary using standard block IO access protocol (eg. SCSI which is Small Computer System Interface). But in the backend, the secondary site in fact use the object storage to store the data. In this way, we can reduce the cost needed to provide a DR solution.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 01 of 5

A disaster recovery solution combining with block and object storage

Background

In the disaster solution of data center, it is typical to configure multiple sites to contain the same data

Sometimes the cost is a big problem
-For high end block storage which is targeted for OLTP task (On Line Transaction Process), it is expensive to prepare two sites to be standby for disaster recovery for some customers

So some customer may choose to use some cheaper storage as the backup or copy site (eg. Object storage) So we need a solution to support this hybrid disaster recovery solution
-The object storage (which is often on cloud) is a typical option The problem of the existing method is
-Efficiency slow, need extra meta data to record the mapping from volume to objects
-Need extra investment on primary site, the primary site is NOT a standard block storage
-Can NOT swap secondary object storage when the primary block storage fails

Here is the basic description of this solution

Organize the key of the value for one block of one volume in this way

    -From the volume ID and block address, we can determine the key for its corresponding object and vice verse. Introduce "Alias Volume" on the secondary site
-The "Alias Volume" is used to establish copy service relationship with the Volume on primary site
-Backend, the data of the "Alias Volume" is organized as object storage
The block and object storage is backup each other


-The details is hidden from the users

    -All update is done on the secondary object storage side
Benefits


-No extra metadata needed


-No special driver on host side

1


Page 02 of 5

-From the user's perspective, they are using two block storage to do disaster recovery

Overall Architecture

Architecture Notes

There is no update needed in Host Server Site

    -It only needs to support standard SCSI drive and provides the swap capability
All changes are in the storage site


-Alias Volume

2


Page 03 of 5

    This the a virtual volume From the host's perspective, it is just a standard SCSI volume whi...