Good Morning --

Our State Computing Services is looking for some assistance.  They are 
troubled by the lack of certainty as to when the data will actually be 
destroyed (i.e. written over) in their current process and are trying to 
develop options for how best to manage these records.  Various options 
were discussed in a meeting yesterday with them, but they are interested 
in seeing how other computing services are handling the matter.  Here is 
the scenario:

Our IBM Mainframe computer service creates tapes that are stored off-site 
for disaster recovery purposes.  The tapes contain copies of valid files 
from the mainframe that would be needed in the event of a fire or other 
physical disaster.  When these tapes no longer have valid data they are 
placed in a “scratch pool” for re-use.  The tapes still contain the 
original data even though the computer system no longer considers them 
valid.  Scratch tapes are randomly selected by the system hardware for 
re-use; theoretically it could be months, years and longer before the data 
on the tapes is written over.  We believe our practices are similar to 
those of other enterprises who operate mainframe computing environments. 
 
Questions: 

How are other computing services managing the data lifecycle on their 
tapes in the above scenario?  Any additional information as to cost, 
employee time, quantity of tapes, etc. - would be helpful.  How do your 
records retention schedules document similar situations? 

Thanks in advance for any/all responses. 

Mary R. Ott, Records Information Management Specialist 
Secretary of State - Record Management Division 
440 South 8th Street, Suite #210
Phone: (402) 471-4184  Fax: (402) 471-2406
Email  [log in to unmask]