1

Determine if data center personnel run the IBM EREP program (IFCEREP1) for error analysis at least once a week.

2

Determine if data center personnel have a program of periodic preventive maintenance (PM) for equipment.

3

Verify that tapes are periodically cleaned, degaussed, and recertified. Does the data center keep maintenance logs, add dates to tape labels, and document tape maintenance?

4

Are high‑maintenance devices (such as tape and cartridge drives, printers, and so on) cleaned in accordance with manufacturer recommendation? This usually involves cleaning drive capstans, heads, and pinch rollers with cotton swabs and alcohol.

5

Determine if follow‑up studies were made to determine the actual cost effectiveness of alternative hardware vendors by analyzing the cost of reruns and down time.

6

From the display, determine if the error rates for tape drives are reasonable and acceptable for the data center. Determine if these rates are monitored on a historical basis against baseline performance levels.

7

Determine if disk drive error rates are reasonably low. Modern disk drives should rarely encounter errors that affect data or applications. Make sure reports on hardware environment status are monitored and preventive maintenance schedules are adjusted when needed so that disk errors, memory parity errors, and so on are addressed while the hardware and operating system can correct them.

Determine whether correctable problems were allowed to escalate into more serious errors that cause job reruns, performance degradation, or service outages.

8

Determine how service level objectives are decided. Make certain they reflect the actual needs of the company. These can include considerations such as:

• Availability objectives (scheduled up‑time hours, infrequency of unavailability, and rarity of teleprocessing or other disconnections, interruptions, or outages.)
• Time‑of‑day performance commitments (response time, turn‑around time, run time, and mount time)
• Capacity headroom objectives (online work, storage, and spool DASD space; print capacity and free CSA)

In this step, concentrate on the hardware vendor and operations responsibilities. Review whether they were met in the past.

9

Are abends, outages, and system down times reported, reviewed, and subjected to follow‑up?