Hi Dwight,
Have to agree there - we have just been given the news that our
ADSM server on MVS has been 'capped' from using too much CPU
cycles or RPP's as they are called lately. This has been decided because
the MVS system ADSM runs on, is part of the production sysplex and thus
causes performance problems on our online production systems.
We have three other environments running WinNT with ATL's attached
to them and I have not had any hassles at all. I'm currently considering to
move the MVS ADSM server to our SP/2. Before server to server comms
was possible this was out of the question but it is looking better by the
minute.
Cheers
Christo
> I have to agree... we have 15 adsm environments 13 aix and 2 mvs
> The battles over cpu cycles on MVS have forced us to eliminate the MVS
> adsm.
>
> Dwight
>
>
>______________________________ Reply Separator
_________________________________
>Subject: Re: transfer speed
>Author: lipp at unix,mime/DD.RFC-822=lipp AT STORSOL DOT COM
>Date: 2/13/99 4:37 PM
>
>
>Mark,
>
>In general, I guess I agree with you. Howsomeever, it seems ADSM in the
MVS env
>ironment defies the rules.
>
>Stick to NT, or UNIX and your costs, and headaches, are probably
substantially l
>ower.
>
>I'm going to steal the cost/pound stuff from you. I think it works!
>
>Kelly J. Lipp
>Storage Solutions Specialists, Inc.
>www.storsol.com
>lipp AT storsol DOT com
>(719)531-5926
>
>-----Original Message-----
>From: Roger Hohmann
>Sent: Friday, February 12, 1999 1:41 AM
>To: ADSM-L AT VM.MARIST DOT EDU
>Subject: Re: transfer speed
>
>Some additional thoughts:
>
>In general, I would agree to minimize the number of systems, but ..
>with adsm it's little different: Some shops have definded downtime
>windows for maintenance (HW and SW). These times are usually at night
>or weekend. This may collide with your backup window and may force you
>to decide between reducing service downtimes, buy another system or
>think about your backup strategy.
>
>In addition, IBM could not deliver (maybe it's different with OS/390
>2.7) a fast tcpip stack, so the performance of those scalable systems is
>limited in ip appls. I can install a ip stack for each ip escon
>connection and a adsm for each ip stack, but then I have multiple adsm
>servers on a single cpu, and that does not minimize administration. So,
>in my opinion, adsm is not very scalable on mvs compared with other
>applications.
>
>And- I expect the difference of costs may come from hw and sw, too. I
>don't expect to pay a million $ for a high end NT box, but a 9672-R26
>is not far away (without disks!). Tape drives and libraries are
>expected to be same price for all platforms. (Don't tell me you have
>some for mvs, your ADSM guy gets them and don't give them back. So you
>have to buy new ones for batch processing and SMS) For a M$, I can pay
>2 guys for 3 yrs. If you upgrade a given S390, you have additional
>costs for sw. You pay your cics and ims and db2 for all cpus in the
>box, not just the used ones. And ask your mvs guys what they think
>about an additional S390, and ask your Unix and NT guys and compare the
>answers.
>
>Scaling an adsm box is not as easy as saying "I have a mvs box, let's
>use it." I payed for it, and I think others payed, too.
>
>Roger
>
>-----Original Message-----
>From: Mapes, Mark [SMTP:MWM4 AT PGE DOT COM]
>Sent: Thursday, February 11, 1999 7:55 PM
>To: ADSM-L AT VM.MARIST DOT EDU
>Subject: Re: transfer speed
>
>Well well well, a discussion that has caught my interest.
>Unfortunately, we can't do a comparison of apples to apples,
>but I have a theory that all computers cost the same per pound (or
>kilogram
>for our non-USA friends).
>So, given an NT box that weighs the same as an AIX box that weighs the
>same
>as an MVS box,
>(and with the assumption that you have the latest and greatest
>hardware/software for each platform, having similar applications and
>capabilities, i.e.
>100mbps Ethernet, etc.)
>Which one will be the better performer? I do not know the answer.
>I can see potential and various advantages that each system may have.
>An NT system has the lowest threshold costs, but is not as scaleable,
>and
>probably has a lower level of functionality (at least in the TCP/IP
>arena).
>An AIX (or generic UNIX) system can probably be implemented near the
>cost of
>a mid to high-tier NT system, will probably perform better and have a
>higher
>level of functionality (again in the TCP/IP arena, but would suffer
>somewhat
>in a pure NT network environment).
>An MVS system would be the most scaleable, be able to support a higher
>amount of simultaneous discrete functions (i.e. batch processing AND
>transaction processing AND data-mining, etc., not necessary ADSM
>related).
>Of course the right answer for you would be some type of vision thing.
> That
>is, where do you want to go to.
>Assuming that the goal is the lowest TCO (total cost of ownership), you
>should be interested in reducing the big costs.
>What are the big costs? Well, I suspect it is not a big cost
>difference of
>buying an NT or UNIX of MVS (again assuming the hardware weighs the
>same).
>Nor would I suspect the cost difference be big between the hardware
>(again
>assuming the hardware weighs the same).
>So what other costs would there be? People!!!
>So, other than paying people less, how do you keep people costs down?
>You minimize the amount of people you need.
>How do you do that? Well, an educated guess (I was taught this is
>school
>many years ago) was that most people can only manage/perform 5 (or is
>it 7
>or 9???) plus or minus 2 tasks effectively. More than that and they
>get
>overwhelmed. (under that they get bored?).
>If each server is considered a task, it would make sense then to
>minimize
>the need for multiple (many) servers.
>Thus, I would tend to think that an MVS system, that is much more
>scaleable
>for all your applications, would be a better choice (with the
>assumptions
>that were made earlier).
>Just some thoughts.
>
>Mark Mapes
>PG&E
>
>
>
>
>> ----------
>> From: Richard Sims[SMTP:rbs AT BU DOT EDU]
>> Reply To: ADSM: Dist Stor Manager
>> Sent: Thursday, February 11, 1999 8:45AM
>> To: ADSM-L AT VM.MARIST DOT EDU
>> Subject: Re: transfer speed
>>
>> >The money spent trying to make that environment fast, IMHO, would be
>> better
>> >spent getting away from that environment. Big NT iron is pretty
>cheap at
>> >this point.
>>
>> Indeed, it has always been an expensive struggle to get mainframes to
>> serve at speeds proportional to their expense.
>> I could only add that big linux iron is even cheaper than NT, and
>> becoming a mission-critical platform for many companies (who are
>> uncomfortable
>> depending upon a monopolistic software supplier).
>> I've heard rumors of IBM coming to support linux, and perhaps we
>> will see an ADSM server for it. Linux has made tremendous strides in
>> the past year, with an ever-increasing amount of commercial software
>> becoming available for it.
>> Richard Sims, BU
>>
>
|
