Brian Peterson wrote:

Hmmm....  I have MEMLIMIT(00006G) coded in SMFPRMxx.  Here's a few entries
from SDSF:

NP   JOBNAME  it StorCrit RptClass MemLimit
    *MASTER*    NO       STCDEF    16383PB
    PCAUTH      NO       PCAUTH
    XCFAS       NO       XCFSTC    16383PB
    GRS         NO       GRSSTC       64PB
    SMSPDSE     NO       STCDEF
    LLA         NO       LLA           6GB

The address spaces which SDSF displays as 16383PB correspond to NOLIMIT in
Mark Zeldon's REXX exec, and address spaces which SDSF displays as (blank)
under MemLimit correspond to 0M in the REXX exec.

The maximum theoretical 64-bit address space is 16 exabytes (EB). Mathematically, that *should* be 16,384 petabytes (PB) or 16,777,216 terabytes (TB). (For those of you that prefer SI notation, please mentally substitute EiB, PiB, TiB, etc.)

In (E)JES I see the following values, depending on how wide I make the MemLimit column:

|Cmd JobName  ACPU%  MemLimit ExCP-Cnt
|<-- --------/------ -------- ---------
|    *MASTER*    .08 16,384PB    11,135
|    PCAUTH      .00                 16
|    RASP        .01                  2
|    TRACE       .00                 60
|    DUMPSRV     .00 16,384PB     5,345
|    XCFAS       .19 16,384PB   502,728
|    GRS         .07 65,536TB        24

|Cmd JobName  ACPU%  MemL ExCP-Cnt
|<-- --------/------ ---- ---------
|    *MASTER*   1.06 16EB    11,136
|    PCAUTH      .00             16
|    RASP        .01              2
|    TRACE       .00             60
|    DUMPSRV     .00 16EB     5,345
|    XCFAS       .24 16EB   502,784
|    GRS         .07 64PB        24

|Cmd JobName  ACPU%  MemLimit     ExCP-Cnt
|<-- --------/------ ------------ ---------
|    *MASTER*    .06 16,777,216TB    11,136
|    PCAUTH      .00                     16
|    RASP        .01                      2
|    TRACE       .00                     60
|    DUMPSRV     .00 16,777,216TB     5,345
|    XCFAS       .35 16,777,216TB   502,784
|    GRS         .03 67,108,864GB        24

It's strange how SDSF agrees that the GRS limit is 64PB but disagrees about 16EB being 16EB. (Could they be using floating point for this?) Though it may seem like a small error, 16,383PB is actually a *lot* less than 16EB. A one PB difference is 1,024TB or 1,048,576GB!

Maybe the answer is for everyone to just to report NOLIMIT like Mark Z does. Then we'll all be in agreement. ;-)

--
-----------------------------------------------------------------
| Edward E. Jaffe                |                                |
| Mgr, Research & Development    | [EMAIL PROTECTED]    |
| Phoenix Software International | Tel: (310) 338-0400 x318       |
| 5200 W Century Blvd, Suite 800 | Fax: (310) 338-0801            |
| Los Angeles, CA 90045          | http://www.phoenixsoftware.com |
-----------------------------------------------------------------

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

Reply via email to