Storage

Most storage administrators didn’t choose that profession; recent headcount reductions have often shifted the responsibility to other technical support staff members. Often, storage administrators have had other primary jobs; they haven’t had the luxury of taking time from their daily job to concentrate on training and learning a new skillset.

But storage won’t just manage itself. A newly hired or newly appointed storage manager can soften the impact of doing more with less and can help a company work smarter. Planning and organization skills can help the new storage manager make the transition less painful, increase productivity, and gain measurable results. Time invested at the beginning of the new job can have significant payout in the coming months. This article presents several steps a new storage manager can take to become more productive in the OS/390 or z/OS environment.

1.       Obtain a copy of the previous storage manager’s JCL library. This library will have most of the jobs, setup utilities, and prewritten procedures needed to successfully manage the new environment.

2.       Track down and obtain a copy of the Standards and Procedures Guide for Storage Management. The documentation will be useful in making modifications to existing procedures such as System-Managed Storage (SMS), Asynchronous Communications Serverroutines, Generation Data Group (GDG) builds, Volume Table of Contents (VTOC) and VSAM Volume Data Set (VVDS) sizing in the shop, storage migration schedule start and end times, restoring data sets, backing up and copying data sets, SMS Management Class retention information, and Hierarchical Storage Management (HSM) information. If this manual isn’t currently available, start organizing one. The manual will benefit you during offhour support issues and will be your salvation when you later try to get in some vacation time.

3.       Obtain mainframe access immediately. Almost all organizations provide offsite access to the mainframe. It’s a necessity to carry both a cell phone and a pager. Time is of the essence when storage problems occur. Most companies will reimburse their employee for DSL service, a pager, and a cell phone. The financial impact of downtime, or of slow response time created by storage problems, easily offsets the expense.

4.       Gain the knowledge necessary to do your job well. Education classes, hands-on training, conferences, and trade magazines are the best means of gaining knowledge. Mainframe hardware and software vendors are a good source for this education. There are also companies specializing in this field. At a minimum, take a basic Storage 101 class covering storage-related system software, and vendor education on the disk hardware and software used at the site. If you’ll also be responsible for the tape system, you must understand the hardware and software involved there, too. This tape knowledge might need to encompass managing hardware such as silos, virtual systems, and extra dense drives. You must understand all the hardware and software related to your storage systems.

Networking with other storage managers is invaluable. “Picking the brain” of existing, long-term storage managers is an important way to gain knowledge. This keeps you, as a new manager, from re-inventing the wheel. When you try to absorb the knowledge of others, you’ll be amazed by the backgrounds of those who have been in the OS/390 or z/OS world for a long time. At some time in their careers, most have been in the same position as any other novice storage manager. Network locally with other companies through any introductions you can arrange, and participate in storage user groups. Most issues you’re facing are similar to those solved by others in the field.

Be sure your reading list includes these magazines and periodicals: z/Journal, Cheryl Watson’s TUNING Letter, Information Week, and ComputerWorld. IBM once published a manual titled How to Lead an Effective Storage Management Group. It was excellent; but it’s no longer in print. Additional sources from IBM include the Redbook series. All IBM information can be accessed through their Website.

5.       Automate your storage procedures. Although most mainframe shops have started tasks to perform storage management, some don’t. If a batch process controls the storage management system, ensure the jobs are executed through a scheduling product. Automate other storage procedures such as capacity planning, storage volume reporting, storage migration activities, backup activities, and recall activities. HSMs have these reporting functions built into the product, but most require manual effort to extract the information. The reporting functions should be automated for quick, easy analysis.

6.       Evaluate and purchase a capable storage-reporting tool.  I’ve been a storage manager for almost 20 years and have evaluated several dozen storagereporting tools. Most tools can produce the reports you need to manage a storage environment. The biggest issue for these products is scalability. Most tools aren’t capable of scaling up and growing with the storage farms. The reporting tool should be flexible and able to report on the whole DASD farm or just a subset of it. The tool should be capable of drilling down to the storage group, storage volume, and the data set levels. It should be able to execute an action at the data set level, including migrating, recalling, deleting, renaming, backing up, and reporting. The reporting tool must be capable of summarizing and reporting on all storage management activities. The tool must be flexible enough to produce ad hoc reports based on several attributes of the stored data, including data set names, size, blocksize, record format, storage group names, and management class names.

7.       This tool must be fast, easily used, and support both Time Sharing Option (TSO) and ISPF. Mainframe storage reporting should be and remain a mainframe tool. This type of tool will let you access and evaluate the entire mainframe environment in less than 30 minutes daily, and frequently during the day, if necessary.

You should have knowledge of a high-level language such as SAS or Easytrieve. This type of language lets you create solutions that significantly exceed the capabilities of IBM’s ISMF tool. You also should be familiar with a non-disruptive volume mover such as Innovation’s FDRPAS or Softek’s TDMF. These tools are useful for moving entire volumes without impacting data use or the rest of the environment. No standalone time is needed to accomplish the entire volume move.

Depending on the hardware you use at your shop, you’ll get much value from EMC Control Center or Catalog Solution, IBM utilities, and DFDSS, to name a few. You’ll find that proficiency in some of the Microsoft software products, such as Excel, is invaluable, and will give you the power to help express yourself clearly and graphically.

A challenge you may encounter in today’s environment is gaining a thorough understanding and mastery of the many iterations of Redundant Array of Independent Disks (RAID) technology, as offered by the various hardware vendors, and what each can provide in your shop. Knowledge of RAID-1, RAID-S, RAID-5 and RAID-10 technology and the pros and cons of each can make your organization a success or failure. Further knowledge of the hardware’s storage internals, such as caching, disk adapters and hard drives, will help you when making hardware purchase decisions.

Don’t forget about the mirroring capabilities of the storage currently on the market. Be sure the hardware you choose will accommodate current and future needs. Don’t buy all the “bells and whistles” unless you plan to use them. Sometimes, you can have unused extras on hardware that can cause vendor microcode problems in your storage complex; you don’t need a “Cadillac solution for a Volkswagen implementation.”

8.       Join a user group. Knowledge sharing and networking within various user groups can be beneficial. IBM’s SHARE, SAS Users Group International (SUGI), and the Computer Measurement Group (CMG) are just a few of the support groups available. Members of these organizations are willing to share information freely on storage management procedures and methods. Some of these organizations require membership dues, which your employer will probably pay. IBM and EMC each have storage symposiums that meet annually and cover current and future storage technologies. Since these are two prominent vendors, it’s wise to arrange to attend their symposiums when possible. You should also attend customer presentations in which presenters often share storage solutions they developed internally to meet their shops’ specific challenges.

9.       Improve your communication skills. Since you need to interact with user groups and higher management, good communication is a requirement. If this is a weak area for you, join a support group geared toward improving your communication skills. Toastmasters International is an example. Effective communication is a skill you must constantly hone anyway, so time spent here is useful no matter which direction your career takes.

As a first-time storage manager, be sure you’re ready to enjoy learning a whole new world. Everyday challenges and rewards will quickly become apparent. The longer you explore this field, the more valuable you’ll become in managing any storage complex. If you don’t learn something new each day, you need to keep digging to advance your knowledge. Welcome to the world of storage managers. Enjoy!