You are currently on IBM Systems Media’s archival website. Click here to view our new website.

MAINFRAME > Storage > Software

Storage Control

How IBM is using DFSMShsm to exploit Parallel Sysplex architecture in the mainframe.


Data storage needs are growing exponentially, as is demand for faster stored data retrieval. These demands are turning the storage administrator's job into a time-sensitive position. How can companies meet these data storage demands in an efficient and cost-effective way? Turn to IBM for the answers. Enhancements to Data Facility Storage Management Subsystem (DFSMS) Hierarchical Storage Manager (HSM) - DFSMShsm* - are helping provide customers with the ability to trim costs and experience superior performance.

DFSMS operates within OS/390* and z/OS*. It's a policy-based storage-management system that uses policies established by the customer. This is designed to enable the system to handle many of the storage-management tasks that were previously done manually. The addition of HSM to DFSMS adds policy-based space management and availability management functions to the system. By utilizing storage group and management class attributes or policies, DFSMShsm is designed to provide automation for both space management and availability management.

The automation of tasks that previously were done manually certainly streamlines the customer's system. Yet, as businesses become more data-centric, quick retrieval of stored data is important. How can companies retrieve necessary data with the fast speed required by today's business environment? IBM believes significantly better performance can be achieved by utilizing a parallel system architecture.

IBM began looking at Parallel Sysplex architecture in the mid 1990s. Pairing DFSMShsm with Parallel Sysplex architecture is one IBM* solution that's designed to provide performance benefits for customers. Enhancements to DFSMShsm have exploited the power of Parallel Sysplex architecture and allowed customers to reap the rewards that come from serving end users' needs quickly and efficiently.

The enhancements, such as the Common Recall Queue (CRQ) and Record Level Sharing (RLS), were developed in response to customer needs.

"We have a process within IBM where we get requirements from customers," says Edward A. Baker, software engineer for IBM Systems and Technology Group, DFSMShsm development. "When we see the same requirement coming in two or three times, we know there's a trend. For instance, we had customers who were looking for better performance when accessing their DFSMShsm control data sets. RLS was developed to address that need."

CRQ was another enhancement spurred by customer requirements.

"On an annual basis, we cull through the hundreds of requirements we have and look for the most frequently requested functions. Both RLS and CRQ were at the top of the list," says Baker.

Enhancements to DFSMShsm have exploited the power of Parallel Sysplex architecture and allowed customers to reap the rewards that come from serving end users' needs quickly and efficiently.

Shirley S. Savage is a Maine-based freelance writer. Shirley can be reached at savage.shirley@comcast.net.



Advertisement

Advertisement

2019 Solutions Edition

A Comprehensive Online Buyer's Guide to Solutions, Services and Education.

A New Storage-Management Strategy

IBM looks to unify mainframe storage space

Disaster Recovery x 3

New TPC for Replication release provides an additional layer of backup

IBM Systems Magazine Subscribe Box Read Now Link Subscribe Now Link iPad App Google Play Store
Mainframe News Sign Up Today! Past News Letters