<snip>
But, now I have a question for the list at large: if Steve does this, is he
in danger of encroaching on shared segments like CMS or PIPES? It doesn't
seem to me that you can safely assume that all storage above 16M is unused
and available.
</snip>
Yes this is a problem, but it is documented in one of the VM manuals. If the virtual machine is defined with 20 meg, storage is layed out one way. If it is defined with storage greater than 20meg then it is layed out a little differently.
I think the shared/saved/NSS/DCSS segments is a dead issue.
We have also discussed the idea of using Virtual Disk and mapping a minidisk to extended storage. XC and all that entails.
We still have to keep in mind how each of these methods would be implemented and how much the change would impact the program(s).
We're trying to make minimal changes to existing programs. i.e., The logic for table processing is already in the programs. Where the table resides, either above the line or below it are inconsequential. Changing the program logic to use mdisks, be it v-disk or dataspaces requires a lot more program change.
Nothing has been ruled out yet, with the exception of the saved segment idea.
So, thanks so far for all of the suggestions and ideas.
I'll let you know how this turns out. But this doesn't mean that this thread has to end here and now.
Steve
| Bill Stephens <[EMAIL PROTECTED]>
Sent by: VM/ESA and z/VM Discussions <[email protected]> 01/13/2006 09:30 AM
|
To: [email protected] cc: Subject: Re: VM maclib reference |
Steve Gentry wrote:
"Can I correctly assumed that Shared Segments (or Saved Segments or NSS or
DCSS) is not an option then? "
-- Probably - you'll be writing into the shared segment, which means that
you won't be able to share the segments with other users, because , iirc,
once you modify a page in a shared segment, your virtual machine gets
"ownership" of that storage and anyone else loading the same segment gets
a pristine (not updated) version of the table. Plus, there's some
administration required to define and instantiate the shared segment.
Because the table is designed to grow, NUCXLOAD doesn't seem to be a good
solution either, because if you overrun the end of your table, you could
possibly overlay other things that got NUCXLOADed behind it.
Now that I have a better understanding of what you need to do, I think I
would proceed with loading the storage above 16M as you stated in your
original post. Yes, you will have to increase the size of the PGMA-running
virtual machines because all storage will have to be contained within them.
But, now I have a question for the list at large: if Steve does this, is he
in danger of encroaching on shared segments like CMS or PIPES? It doesn't
seem to me that you can safely assume that all storage above 16M is unused
and available.
Regards,
Bill Stephens
Sr. Technology Analyst, High Availability
SunGard Availability Services
10th floor
401 North Broad Street
Philadelphia, PA 19108
Phone: (215) 351-1099
Fax: (215) 451-2045
[EMAIL PROTECTED]
___________________________________________
Keeping People and Information Connected (TM)
http://www.availability.sungard.com
Steve Gentry
<Stephen_R_Gentry
@LafayetteLife.co To
m> [email protected]
Sent by: VM/ESA cc
and z/VM
Discussions Subject
<[EMAIL PROTECTED] Re: VM maclib reference
.UARK.EDU>
01/13/2006 08:16
AM
Please respond to
VM/ESA and z/VM
Discussions
<[EMAIL PROTECTED]
.UARK.EDU>
They will be updated as the program runs. When PGMA starts, all tables are
empty (null, whatever). As the program runs, it starts filling up the
table starting at the
beginning of the table. Data is continually added(or updated) to the table
during the life of PGMA.
Can I correctly assumed that Shared Segments (or Saved Segments or NSS or
DCSS) is not an option then?
Bill Stephens <[EMAIL PROTECTED]>
Sent by: VM/ESA and z/VM Discussions To:
<[email protected]> [email protected]
cc:
Subject:
01/13/2006 08:02 AM Re: VM maclib reference
Please respond to VM/ESA and z/VM
Discussions
Steve Gentry wrote:
"The other thought has been to use Shared Segments. I need some
clarification on this technique. The program involved (let's call it PGMA)
can be invoked by multiple users at the same time. The table will probably
contain different data for each user depending on what they are trying to."
-- Well, that probably leaves out shared segments. I was under the
impression that all users would be sharing the same data; I actually worked
on an application like this decades ago, unfortunately don't remember many
details other than it could be done.
"Will I need to define a Segment for each user or when PGMA is invoked will
it get its own unique area of storage to load the table? My thinking is
that it will not get a unique chunk of storage for each user. If this is
the case, shared segments is not an option."
-- I would think you would need a shared segment for each table; this way,
more than one user could use the same table. The segment addresses could
overlap if each user uses only one table; otherwise the shared segments
will all have to be at different addresses to allow multiple tables to be
loaded as needed.
-- Are these static tables (e.g., used for data lookup), or will they be
updated?
Regards,
Bill Stephens
Sr. Technology Analyst, High Availability
SunGard Availability Services
10th floor
401 North Broad Street
Philadelphia, PA 19108
Phone: (215) 351-1099
Fax: (215) 451-2045
[EMAIL PROTECTED]
___________________________________________
Keeping People and Information Connected (TM)
http://www.availability.sungard.com
Steve Gentry
<Stephen_R_Gentry
@LafayetteLife.co To
m> [email protected]
Sent by: VM/ESA cc
and z/VM
Discussions Subject
<[EMAIL PROTECTED] Re: VM maclib reference
.UARK.EDU>
01/13/2006 07:53
AM
Please respond to
VM/ESA and z/VM
Discussions
<[EMAIL PROTECTED]
.UARK.EDU>
There have been two avenues of thought here. a) use CMSSTOR and RMODE and
AMODE, etc, to load the table above 16meg. In doing this I think I will
also have to make the users virtual machine size larger than 16meg, right?
The other thought has been to use Shared Segments. I need some
clarification on this technique. The program involved (let's call it PGMA)
can be invoked by multiple users at the same time. The table will probably
contain different data for each user depending on what they are trying to.
Will I need to define a Segment for each user or when PGMA is invoked will
it get its own unique area of storage to load the table? My thinking is
that it will not get a unique chunk of storage for each user. If this is
the case, shared segments is not an option.
Could someone explain this?
Please and thank you.
Steve
