Browse Prior Art Database

Method for Continuous Access to Shared Removable Media and Devices from Multiple Nodes

IP.com Disclosure Number: IPCOM000146223D
Original Publication Date: 2007-Feb-08
Included in the Prior Art Database: 2007-Feb-08
Document File: 3 page(s) / 79K

Publishing Venue

IBM

Abstract

Provided is a method for managing a library of removable media and devices shared by multiple Virtual Tape Server (VTS) nodes in a storage network. The method is implemented as a computer readable program, with an instance running on each VTS node in the network, having the common knowledge of the availability and accessibility of the library's physical resources. A common database shared by each VTS node retains information about library resources and table "locks" that control access to those resources. In addition, the VTS node requiring media input/output access to a device must inquire the database and reserve the resource at both the database and device levels.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 47% of the total text.

Page 1 of 3

Method for Continuous Access to Shared Removable Media and Devices from Multiple Nodes

Disclosed is a method to manage a pool of physical library resources shared by two Virtual Tape Server nodes capable of: (1) communicating with one another in a storage network, (2) sharing and synchronizing database information between the nodes, (3) accessing a plurality of removable media and devices sequentially, and (4) reporting and monitoring node status to facilitate an automatic release of one or more resources owned by the failed node.

The figure below is a high level block diagram showing the Applicants' dual node data storage system; the VTS nodes are interconnected via a TCP/IP network; the IBM DB2 database management system on each node provides the High Availability Disaster Recovery (HADR) feature, which allows one node to communicate with a second node via the TCP/IP network to share database information; each node is physical connected to tape drives via fiber channel switches.

The prior art virtual tape server is a single node storage management subsystem coupled to a tape library consisting of a plurality of tape cartridges and devices. This virtual tape server is the sole owner of the tape cartridges and devices that belong to the subsystem. A second virtual tape server cannot be connected to the same tape library partition to co-manage the same tape cartridges and devices. Hence, if the prior art, single node, virtual tape server is taken out of service due to a scheduled maintenance or problem in the subsystem, the down time may be disruptive to the customer's business continuity.

The prior art peer-to-peer VTS solution provides customers with improved business continuity by copying data to both peered systems such that if one undergoes maintenance or encounters a

1

[This page contains 2 pictures or other non-text objects]

Page 2 of 3

problem, customer data can be retrieved from and written to the active system. A limitation of this peer-to-peer storage system is that if one side is down, the means of offsite data redundancy is deferred until the inactive system is brought online.

With the above limitations, there is a need for better techniques and systems that provide greater availability to customer data stored on removable media, regardless of whether or not the storage system is a peer-to-peer VTS. Hence, the details of a better method is described.

The method will be implemented as a computer readable program code residing on each Virtual Tape Server node. The program is called to start when the node is instructed to go online, establishing its initial state in the subsystem. As part of the initialization process, the method will gather information about the VTS subsystem, including whether the system is a single or dual node, which physical devices are configured to the system, and the status of e...