[UR-WG] LAST CALL - Usage Record Format Recommendation - Version 1

Laura F McGinnis lfm at psc.edu
Fri Sep 22 13:46:10 CDT 2006


I have updated Appendix C to note the differences between SI prefixes and
binary prefixes. Rather than mandate one over the other, I have added the
text: 

"Usage record measures of volume should use the appropriate terminology for
referring to units. For measurements based on powers of 2 (2^10, 2^20, etc),
the prefixes for binary multiples should be used. For measurements based on
powers of 10 (10^3, 10^6, etc), the SI prefixes should be used."

I have also included charts from NIST, which are based on the IEC standards.

If this is not sufficient, please let me know.

Thx
LM

> -----Original Message-----
> From: ur-wg-bounces at ogf.org [mailto:ur-wg-bounces at ogf.org] On 
> Behalf Of Donal K. Fellows
> Sent: Thursday, September 21, 2006 6:40 AM
> To: Mailing List for UR-WG
> Subject: Re: [UR-WG] LAST CALL - Usage Record Format 
> Recommendation - Version 1
> >>  * Appendix C should note that SI usage requires "kiB" for 
> >>1024 bytes,
> >>    "MiB" for 1024 kiB, etc. It's stupid, but "standard". 
> It's also a
> >>    large part of the reason why JSDL dropped putting units in. :-\
> > 
> > What is "SI Usage"? How should this be noted in Appendix C? 
> 
> Apparently, we're supposed to use 1MB to refer to 1000000 
> bytes and 1MiB
> to refer to 1048576 bytes (== 1024kiB, 1kiB == 1024 bytes). I 
> hate this
> usage, but it is standard. http://en.wikipedia.org/wiki/Binary_prefix
> has lots of gory details, including links to more discussion.
> 
> I prefer to put everything in bytes or bits and use very big 
> numbers. :-)
 
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Laura F. McGinnis, Project Coordinator
Data & Information Resource Services
Pittsburgh Supercomputing Center                      email: lfm at psc.edu
300 South Craig St, #313                             phone: 412-268-5642
Pittsburgh, PA  15213                                  fax: 412-268-5832
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 



More information about the ur-wg mailing list