Re: FLASHCOPY performance to a DS6800 DASD

2008-07-16 Thread Macioce, Larry
I can't help you with the flashcopy question, but I can tell you our
dfdss full backups(3390-3) went from 20-25 min to 6-7 min when we
changed out to the ds6k.

No one could believe it

 



From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On
Behalf Of Edward M. Martin
Sent: Wednesday, July 16, 2008 10:00 AM
To: IBMVM@LISTSERV.UARK.EDU
Subject: FLASHCOPY performance to a DS6800 DASD

 

Hello Everyone,

 

 I need some confirmation.  I am just amazed at the performance of
the DS6800 

 

 I just did the z/VM command FLASHCOPY on a 3390-mod 9 on DS6800 to
another 3390-mod 9 volume on the DS6800.

 

FLASHCOPY 8A2 0 END TO 8A1 0 END SYNCHRONOUS

 

 The command responded instantaneously.

 

SYNChronous


tells CP to process the command immediately and does not allow you
to

enter any other commands until the hardware has accepted all
parameters, 

all messages from the Enterprise Storage Server (ESS) subsystem have
been

processed, and the FLASHCOPY command completes. 

 

Is the DS6800 and the FLASHCOPY really that fast or is there something
going on behind the 

scenes?

 



Performance has been pretty unbelievable but the FLASHCOPY was just too
fast.

 

Example Our Batch cycle at night use to take 5-5.5 hours.  Started at
midnight and 

Completeed around 5:30 am.

 

Old system EMC SYMETRIC 8 gig of cache and 6 ESCON connections.

 

SWITCH to DS6800.

 



DS6800 2 gig of cache with 4 FICON connections.

 

Batch cycle dropped by 1 hour without any other changes. Start at
midnight and completes

around 4-4:30 am.   

   

 

 

 

Ed Martin

330-588-4723

ext 40441

 




-

The information transmitted is intended solely for the individual
or entity to which it is addressed and may contain confidential
and/or
privileged material. Any review, retransmission, dissemination or
other use of or taking action in reliance upon this information by
persons or entities other than the intended recipient is
prohibited. If you have received this email in error please contact
the sender and delete the
material from any computer.



Re: FLASHCOPY performance to a DS6800 DASD

2008-07-16 Thread Edward M. Martin
Hello Larry,

 

 I kept hearing those types of numbers.  I am very pleasantly
surprised.

 

Ed Martin

330-588-4723

ext 40441



From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On
Behalf Of Macioce, Larry
Sent: Wednesday, July 16, 2008 10:06 AM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: FLASHCOPY performance to a DS6800 DASD

 

I can't help you with the flashcopy question, but I can tell you our
dfdss full backups(3390-3) went from 20-25 min to 6-7 min when we
changed out to the ds6k.

No one could believe it

 



From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On
Behalf Of Edward M. Martin
Sent: Wednesday, July 16, 2008 10:00 AM
To: IBMVM@LISTSERV.UARK.EDU
Subject: FLASHCOPY performance to a DS6800 DASD

 

Hello Everyone,

 

 I need some confirmation.  I am just amazed at the performance of
the DS6800 

 

 I just did the z/VM command FLASHCOPY on a 3390-mod 9 on DS6800 to
another 3390-mod 9 volume on the DS6800.

 

FLASHCOPY 8A2 0 END TO 8A1 0 END SYNCHRONOUS

 

 The command responded instantaneously.

 

SYNChronous


tells CP to process the command immediately and does not allow you
to

enter any other commands until the hardware has accepted all
parameters, 

all messages from the Enterprise Storage Server (ESS) subsystem have
been

processed, and the FLASHCOPY command completes. 

 

Is the DS6800 and the FLASHCOPY really that fast or is there something
going on behind the 

scenes?

 



Performance has been pretty unbelievable but the FLASHCOPY was just too
fast.

 

Example Our Batch cycle at night use to take 5-5.5 hours.  Started at
midnight and 

Completeed around 5:30 am.

 

Old system EMC SYMETRIC 8 gig of cache and 6 ESCON connections.

 

SWITCH to DS6800.

 



DS6800 2 gig of cache with 4 FICON connections.

 

Batch cycle dropped by 1 hour without any other changes. Start at
midnight and completes

around 4-4:30 am.   

   

 

 

 

Ed Martin

330-588-4723

ext 40441

 



 The
information transmitted is intended solely for the individual or entity
to which it is addressed and may contain confidential and/or privileged
material. Any review, retransmission, dissemination or other use of or
taking action in reliance upon this information by persons or entities
other than the intended recipient is prohibited. If you have received
this email in error please contact the sender and delete the material
from any computer.
 



Re: FLASHCOPY performance to a DS6800 DASD

2008-07-16 Thread Bob Levad (641-585-6770)
Ed,
 
I'm sure I'll be corrected if my understanding is in error.
 
The FLASHCOPY proceeds in the background, but the source is available for
use immediately as writes to un-copied portions of the source volume are
logged and held until the flashcopy completes.  
 
Also, reads of uncopied areas of the destination volume pull from the
original, as yet unchanged, volume.
 
Once the copy is complete, the logged updates are applied to the source
volume.
 
On our DS8100, we flashcopy about 60 VSE volumes each night (waiting 3
seconds between commands) and then immediately bring up our production
system.  We then immediately start VMBACKUP to grab a full DR copy from the
destination volumes.
 
Our production VSE is IPL'd usually within 5 minutes of shutdown.
 
Bob

  _  

From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On
Behalf Of Edward M. Martin
Sent: Wednesday, July 16, 2008 9:17 AM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: FLASHCOPY performance to a DS6800 DASD



Hello Larry,

 

 I kept hearing those types of numbers.  I am very pleasantly surprised.

 

Ed Martin

330-588-4723

ext 40441

  _  

From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On
Behalf Of Macioce, Larry
Sent: Wednesday, July 16, 2008 10:06 AM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: FLASHCOPY performance to a DS6800 DASD

 

I can't help you with the flashcopy question, but I can tell you our dfdss
full backups(3390-3) went from 20-25 min to 6-7 min when we changed out to
the ds6k.

No one could believe it

 

  _  

From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On
Behalf Of Edward M. Martin
Sent: Wednesday, July 16, 2008 10:00 AM
To: IBMVM@LISTSERV.UARK.EDU
Subject: FLASHCOPY performance to a DS6800 DASD

 

Hello Everyone,

 

 I need some confirmation.  I am just amazed at the performance of the
DS6800 

 

 I just did the z/VM command FLASHCOPY on a 3390-mod 9 on DS6800 to
another 3390-mod 9 volume on the DS6800.

 

FLASHCOPY 8A2 0 END TO 8A1 0 END SYNCHRONOUS

 

 The command responded instantaneously.

 

SYNChronous


tells CP to process the command immediately and does not allow you to


enter any other commands until the hardware has accepted all parameters,


all messages from the Enterprise Storage Server (ESS) subsystem have
been

processed, and the FLASHCOPY command completes. 

 

Is the DS6800 and the FLASHCOPY really that fast or is there something going
on behind the 

scenes?

 



Performance has been pretty unbelievable but the FLASHCOPY was just too
fast.

 

Example Our Batch cycle at night use to take 5-5.5 hours.  Started at
midnight and 

Completeed around 5:30 am.

 

Old system EMC SYMETRIC 8 gig of cache and 6 ESCON connections.

 

SWITCH to DS6800.

 



DS6800 2 gig of cache with 4 FICON connections.

 

Batch cycle dropped by 1 hour without any other changes. Start at midnight
and completes

around 4-4:30 am.   

   

 

 

 

Ed Martin

330-588-4723

ext 40441

 


  _  

 The
information transmitted is intended solely for the individual or entity to
which it is addressed and may contain confidential and/or privileged
material. Any review, retransmission, dissemination or other use of or
taking action in reliance upon this information by persons or entities other
than the intended recipient is prohibited. If you have received this email
in error please contact the sender and delete the material from any
computer.
 


This electronic transmission and any documents accompanying this electronic 
transmission contain confidential information belonging to the sender.  This 
information may be legally privileged.  The information is intended only for 
the use of the individual or entity named above.  If you are not the intended 
recipient, you are hereby notified that any disclosure, copying, distribution, 
or the taking of any action in reliance on or regarding the contents of this 
electronically transmitted information is strictly prohibited.



Re: FLASHCOPY performance to a DS6800 DASD

2008-07-16 Thread Edward M. Martin
Hello and Thanks to everyone,

I have been to the lectures and read the literature.  I was just
amazed at the performance.

Question (because I am Systems), what happens to the copy
process if power is interrupted?
And how do I know when the copy function is actually completed?

Ed Martin
330-588-4723
ext 40441

From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On
Behalf Of Bob Levad (641-585-6770)
Sent: Wednesday, July 16, 2008 10:36 AM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: FLASHCOPY performance to a DS6800 DASD

Ed,
 
I'm sure I'll be corrected if my understanding is in error.
 
The FLASHCOPY proceeds in the background, but the source is available
for use immediately as writes to un-copied portions of the source volume
are logged and held until the flashcopy completes.  
 
Also, reads of uncopied areas of the destination volume pull from the
original, as yet unchanged, volume.
 
Once the copy is complete, the logged updates are applied to the source
volume.
 
On our DS8100, we flashcopy about 60 VSE volumes each night (waiting 3
seconds between commands) and then immediately bring up our production
system.  We then immediately start VMBACKUP to grab a full DR copy from
the destination volumes.
 
Our production VSE is IPL'd usually within 5 minutes of shutdown.
 
Bob

ntents of this electronically transmitted information is strictly
prohibited. 


Re: FLASHCOPY performance to a DS6800 DASD

2008-07-16 Thread RPN01
We use it for our Linux image cloning, where we copy two to three 3390 mod 9
volumes to create the new image. From request to first boot of the copied
image is roughly 18 to 25 seconds. CanĀ¹t fault that at all...

-- 
Robert P. Nix  Mayo Foundation.~.
RO-OE-5-55 200 First Street SW/V\
507-284-0844   Rochester, MN 55905   /( )\
-^^-^^
In theory, theory and practice are the same, but
 in practice, theory and practice are different.




On 7/16/08 9:43 AM, Edward M. Martin [EMAIL PROTECTED] wrote:

 Hello and Thanks to everyone,
 
 I have been to the lectures and read the literature.  I was just
 amazed at the performance.
 
 Question (because I am Systems), what happens to the copy process if
 power is interrupted?
 
 And how do I know when the copy function is actually completed?
 
 Ed Martin
 
 330-588-4723
 
 ext 40441
 
 From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf
 Of Bob Levad (641-585-6770)
 Sent: Wednesday, July 16, 2008 10:36 AM
 To: IBMVM@LISTSERV.UARK.EDU
 Subject: Re: FLASHCOPY performance to a DS6800 DASD
 
 Ed,
 
  
 
 I'm sure I'll be corrected if my understanding is in error.
 
  
 
 The FLASHCOPY proceeds in the background, but the source is available for use
 immediately as writes to un-copied portions of the source volume are logged
 and held until the flashcopy completes.
 
  
 
 Also, reads of uncopied areas of the destination volume pull from the
 original, as yet unchanged, volume.
 
  
 
 Once the copy is complete, the logged updates are applied to the source
 volume.
 
  
 
 On our DS8100, we flashcopy about 60 VSE volumes each night (waiting 3 seconds
 between commands) and then immediately bring up our production system.  We
 then immediately start VMBACKUP to grab a full DR copy from the destination
 volumes.
 
  
 
 Our production VSE is IPL'd usually within 5 minutes of shutdown.
 
  
 
 Bob
 
 ntents of this electronically transmitted information is strictly prohibited.
 




Re: FLASHCOPY performance to a DS6800 DASD

2008-07-16 Thread Tom Duerbusch
Hi Ed

Yep, it is amazing, especially for those of us that never had access to a high 
performance dasd subsystem before.

Without the sync parm, the command comes back immediately.  With the sync parm, 
the command comes back when the DS6800 has gathered all the information/control 
blocks necessary, and starts the flashcopy.

I've never been able to determine when a flashcopy is completed, other than 
looking at the lights on the box.  That has been a problem as if you need to 
rerun the flashcopy command again (like something bombed out in the rexx exec 
and it is easier just to reexecute the exec), it tells you that the to volume 
is in use.  So if you have a mixture of large and small volumes, the small 
volumes get a repeat of being flashed and the large volumes end up with the 
original flash.

There is suppose to be a way from the DS6800 console to determine what flashes 
are running, but I never got that far.  Also, if you want to flash a series of 
volumes at the same instant, it can be done via the DS6800 console.  

I flash minidisks, so I can only use the VM Flashcopy command.

BTW, have you noticed that when you run batch jobs, you are always near 100% 
cpu utilization?  My boss keeps asking me who is in a loop.  We are running 
just fine G.

Tom Duerbusch
THD Consulting  

Law of Cat Acceleration

  A cat will accelerate at a constant rate, until he gets good and
  ready to stop.


 Edward M. Martin [EMAIL PROTECTED] 7/16/2008 8:59 AM 


Hello Everyone,

 

 I need some confirmation.  I am just amazed at the performance of
the DS6800 

 

 I just did the z/VM command FLASHCOPY on a 3390-mod 9 on DS6800 to
another 3390-mod 9 volume on the DS6800.

 

FLASHCOPY 8A2 0 END TO 8A1 0 END SYNCHRONOUS

 

 The command responded instantaneously.

 

SYNChronous


tells CP to process the command immediately and does not allow you
to

enter any other commands until the hardware has accepted all
parameters, 

all messages from the Enterprise Storage Server (ESS) subsystem have
been

processed, and the FLASHCOPY command completes. 

 

Is the DS6800 and the FLASHCOPY really that fast or is there something
going on behind the 

scenes?

 



Performance has been pretty unbelievable but the FLASHCOPY was just too
fast.

 

Example Our Batch cycle at night use to take 5-5.5 hours.  Started at
midnight and 

Completeed around 5:30 am.

 

Old system EMC SYMETRIC 8 gig of cache and 6 ESCON connections.

 

SWITCH to DS6800.

 



DS6800 2 gig of cache with 4 FICON connections.

 

Batch cycle dropped by 1 hour without any other changes. Start at
midnight and completes

around 4-4:30 am.   

   

 

 

 

Ed Martin

330-588-4723

ext 40441

 


Re: FLASHCOPY performance to a DS6800 DASD

2008-07-16 Thread Edward M. Martin
Hello Tom,

I have not noticed the 100 during the batch processing but I
will ask the operators to keep an eye on it.

Ed Martin
330-588-4723
ext 40441

-Original Message-
From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On
Behalf Of Tom Duerbusch
Sent: Wednesday, July 16, 2008 11:26 AM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: FLASHCOPY performance to a DS6800 DASD

Hi Ed

Yep, it is amazing, especially for those of us that never had access to
a high performance dasd subsystem before.

Without the sync parm, the command comes back immediately.  With the
sync parm, the command comes back when the DS6800 has gathered all the
information/control blocks necessary, and starts the flashcopy.

I've never been able to determine when a flashcopy is completed, other
than looking at the lights on the box.  That has been a problem as if
you need to rerun the flashcopy command again (like something bombed out
in the rexx exec and it is easier just to reexecute the exec), it tells
you that the to volume is in use.  So if you have a mixture of large
and small volumes, the small volumes get a repeat of being flashed and
the large volumes end up with the original flash.

There is suppose to be a way from the DS6800 console to determine what
flashes are running, but I never got that far.  Also, if you want to
flash a series of volumes at the same instant, it can be done via the
DS6800 console.  

I flash minidisks, so I can only use the VM Flashcopy command.

BTW, have you noticed that when you run batch jobs, you are always near
100% cpu utilization?  My boss keeps asking me who is in a loop.  We are
running just fine G.

Tom Duerbusch
THD Consulting  

Law of Cat Acceleration

  A cat will accelerate at a constant rate, until he gets good and
  ready to stop.


 Edward M. Martin [EMAIL PROTECTED] 7/16/2008 8:59 AM 


Hello Everyone,

 

 I need some confirmation.  I am just amazed at the performance of
the DS6800 

 

 I just did the z/VM command FLASHCOPY on a 3390-mod 9 on DS6800 to
another 3390-mod 9 volume on the DS6800.

 

FLASHCOPY 8A2 0 END TO 8A1 0 END SYNCHRONOUS

 

 The command responded instantaneously.

 

SYNChronous


tells CP to process the command immediately and does not allow you
to

enter any other commands until the hardware has accepted all
parameters, 

all messages from the Enterprise Storage Server (ESS) subsystem have
been

processed, and the FLASHCOPY command completes. 

 

Is the DS6800 and the FLASHCOPY really that fast or is there something
going on behind the 

scenes?

 



Performance has been pretty unbelievable but the FLASHCOPY was just too
fast.

 

Example Our Batch cycle at night use to take 5-5.5 hours.  Started at
midnight and 

Completeed around 5:30 am.

 

Old system EMC SYMETRIC 8 gig of cache and 6 ESCON connections.

 

SWITCH to DS6800.

 



DS6800 2 gig of cache with 4 FICON connections.

 

Batch cycle dropped by 1 hour without any other changes. Start at
midnight and completes

around 4-4:30 am.   

   

 

 

 

Ed Martin

330-588-4723

ext 40441

 


Re: FLASHCOPY performance to a DS6800 DASD

2008-07-16 Thread Kris Buelens
Check out ICKDSF's PPRCOPY command.  I surely has information about pending
remote copies.
   ICKDSF CONSOLE CONSOLE
   PPRCOPY QUERY UNIT(vaddr)
  (you need to link to a fullpack first; you can query the primary and
secondary device)
So, it may also return information about Flashcopy status.

2008/7/16 Tom Duerbusch [EMAIL PROTECTED]:

 Hi Ed

 Yep, it is amazing, especially for those of us that never had access to a
 high performance dasd subsystem before.

 Without the sync parm, the command comes back immediately.  With the sync
 parm, the command comes back when the DS6800 has gathered all the
 information/control blocks necessary, and starts the flashcopy.

 I've never been able to determine when a flashcopy is completed, other than
 looking at the lights on the box.  That has been a problem as if you need to
 rerun the flashcopy command again (like something bombed out in the rexx
 exec and it is easier just to reexecute the exec), it tells you that the
 to volume is in use.  So if you have a mixture of large and small volumes,
 the small volumes get a repeat of being flashed and the large volumes end up
 with the original flash.

 There is suppose to be a way from the DS6800 console to determine what
 flashes are running, but I never got that far.  Also, if you want to flash a
 series of volumes at the same instant, it can be done via the DS6800
 console.

 I flash minidisks, so I can only use the VM Flashcopy command.

 BTW, have you noticed that when you run batch jobs, you are always near
 100% cpu utilization?  My boss keeps asking me who is in a loop.  We are
 running just fine G.

 Tom Duerbusch
 THD Consulting

 Law of Cat Acceleration

  A cat will accelerate at a constant rate, until he gets good and
  ready to stop.


  Edward M. Martin [EMAIL PROTECTED] 7/16/2008 8:59 AM 


 Hello Everyone,



 I need some confirmation.  I am just amazed at the performance of
 the DS6800



 I just did the z/VM command FLASHCOPY on a 3390-mod 9 on DS6800 to
 another 3390-mod 9 volume on the DS6800.



 FLASHCOPY 8A2 0 END TO 8A1 0 END SYNCHRONOUS



 The command responded instantaneously.



 SYNChronous


tells CP to process the command immediately and does not allow you
 to

enter any other commands until the hardware has accepted all
 parameters,

all messages from the Enterprise Storage Server (ESS) subsystem have
 been

processed, and the FLASHCOPY command completes.



 Is the DS6800 and the FLASHCOPY really that fast or is there something
 going on behind the

 scenes?





 Performance has been pretty unbelievable but the FLASHCOPY was just too
 fast.



 Example Our Batch cycle at night use to take 5-5.5 hours.  Started at
 midnight and

 Completeed around 5:30 am.



Old system EMC SYMETRIC 8 gig of cache and 6 ESCON connections.



 SWITCH to DS6800.





DS6800 2 gig of cache with 4 FICON connections.



 Batch cycle dropped by 1 hour without any other changes. Start at
 midnight and completes

around 4-4:30 am.









 Ed Martin

 330-588-4723

 ext 40441





-- 
Kris Buelens,
IBM Belgium, VM customer support


Re: FLASHCOPY performance to a DS6800 DASD

2008-07-16 Thread Steve Wilkins

For those on z/VM 5.3.0, please check out the CP QUERY Virtual FLASHCopy
command;  under FLC
in the online help for HELP CPQUERY VIRTUAL.

especially the ACTIVE Number of active FLASHCOPY
relationships output.   It shows the number of
copies (including background copies) currently active for the real
track extents (start to end) covered by a
VM minidisk on the real volume.

Steve Wilkins
z/VM I/O Strategy


   
 Kris Buelens  
 [EMAIL PROTECTED] 
 il.comTo
 Sent by: The IBM  IBMVM@LISTSERV.UARK.EDU 
 z/VM Operating cc
 System
 [EMAIL PROTECTED] Subject
 ARK.EDU  Re: FLASHCOPY performance to a  
   DS6800 DASD 
   
 07/16/2008 01:37  
 PM
   
   
 Please respond to 
   The IBM z/VM
 Operating System  
 [EMAIL PROTECTED] 
 ARK.EDU  
   
   




Check out ICKDSF's PPRCOPY command.  I surely has information about pending
remote copies.
   ICKDSF CONSOLE CONSOLE
   PPRCOPY QUERY UNIT(vaddr)
  (you need to link to a fullpack first; you can query the primary and
secondary device)
So, it may also return information about Flashcopy status.

2008/7/16 Tom Duerbusch [EMAIL PROTECTED]:
  Hi Ed

  Yep, it is amazing, especially for those of us that never had access to a
  high performance dasd subsystem before.

  Without the sync parm, the command comes back immediately.  With the sync
  parm, the command comes back when the DS6800 has gathered all the
  information/control blocks necessary, and starts the flashcopy.

  I've never been able to determine when a flashcopy is completed, other
  than looking at the lights on the box.  That has been a problem as if you
  need to rerun the flashcopy command again (like something bombed out in
  the rexx exec and it is easier just to reexecute the exec), it tells you
  that the to volume is in use.  So if you have a mixture of large and
  small volumes, the small volumes get a repeat of being flashed and the
  large volumes end up with the original flash.

  There is suppose to be a way from the DS6800 console to determine what
  flashes are running, but I never got that far.  Also, if you want to
  flash a series of volumes at the same instant, it can be done via the
  DS6800 console.

  I flash minidisks, so I can only use the VM Flashcopy command.

  BTW, have you noticed that when you run batch jobs, you are always near
  100% cpu utilization?  My boss keeps asking me who is in a loop.  We are
  running just fine G.

  Tom Duerbusch
  THD Consulting

  Law of Cat Acceleration

   A cat will accelerate at a constant rate, until he gets good and
   ready to stop.


   Edward M. Martin [EMAIL PROTECTED] 7/16/2008 8:59 AM 


  Hello Everyone,



  I need some confirmation.  I am just amazed at the performance of
  the DS6800



  I just did the z/VM command FLASHCOPY on a 3390-mod 9 on DS6800 to
  another 3390-mod 9 volume on the DS6800.



  FLASHCOPY 8A2 0 END TO 8A1 0 END SYNCHRONOUS



  The command responded instantaneously.



  SYNChronous


 tells CP to process the command immediately and does not allow you
  to

 enter any other commands until the hardware has accepted all
  parameters,

 all messages from the Enterprise Storage Server (ESS) subsystem have
  been

 processed, and the FLASHCOPY command completes.



  Is the DS6800 and the FLASHCOPY really that fast or is there something
  going on behind the

  scenes?





  Performance has been pretty unbelievable but the FLASHCOPY was just too
  fast.



  Example Our Batch cycle at night use to take 5-5.5 hours.  Started at
  midnight and

  Completeed around 5:30 am.



 Old system EMC SYMETRIC 8 gig of cache and 6 ESCON connections.



  SWITCH to DS6800.





 DS6800 2 gig of cache with 4 FICON connections.



  Batch cycle