IBM-AUSTRIA - PC-HW-Support    30 Aug 1999

Predictive Failure Analysis

As with any electrical/mechanical device, there are two basic failure types:

1.  Gradual performance degradation  of components can create a catastrophic drive failure.

    Predictive Failure Analysis performs the following remedial  operations:

   •  Monitors performance of drives
   •  Analyzes data from periodic internal measurements
   •  Recommends replacement when specific thresholds are exceeded

    The data from periodic internal measurements is collected when data  sectors are accessed.

    Data scrubbing performs the following operations:

   •  Forces all data sectors to be read
   •  Provides more data to improve the

    accuracy of PFA The thresholds have been determined by examining the  history logs of drives that have failed in actual customer operation.  When PFA detects a threshold exceeded failure, the system administrator  can be notified through Netfinity Manager 5.0. The design goal of PFA is  to provide a minimum of 24 hours warning before a drive experiences  'catastrophic' failure.

2.  A cable breaking, a component burning out, a  solder connection failing, are all examples of 'on/off' unpredictable  catastrophic failures . As assembly and component processes have  improved, these types of defects have been reduced but not eliminated.  PFA cannot always provide warning for on/off unpredictable failures.

Please see the LEGAL  -  Trademark notice.
Feel free - send a   for any BUG on this page found - Thank you.