Keep in mind this thread is about AWS (the magnificent) having an outage, that
if it happened in a mainframe shop, would be considered terrible and
unacceptable. In fact, AWS (and Azure/Google clouds) are down fairly often.
There are people on the internet who keep track of these things. Whereas the
mainframes at large companies in which downtime can cost millions of dollars
and yes even lives almost never goes down. But rest assured, the media would
overemphasize the mainframe outage on the "legacy" platform as if people are
still running mainframe technology from 50 years ago. Because heck, as we all
know, only the new stuff gets investment dollars and is considered 21st century
tech! (LOL) Now prepare yourself for a walk down memory lane. And, this is
likely my last post on this thread. (hurray) Because I have actual work to do
and more pleasurable and relaxing distractions. However, I do reserve the right
to rescind my last statement should I deem fit.
On Thursday, December 9, 2021, 07:00:53 AM EST, Seymour J Metz
<[email protected]> wrote:
Keep in mind that what IBM is discussing is throughput rather than the
duration of a single I/O. A typical installation will be doing I/O to a lot of
DASD volumes concurrently. When you have a lot of concurrent I/O requests on
the same channel, the effect of, e.g., rotational delay, cache miss, is mostly
on duration rather than on throughput.
--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
________________________________________
From: IBM Mainframe Discussion List [[email protected]] on behalf of
kekronbekron [[email protected]]
Sent: Wednesday, December 8, 2021 10:52 PM
To: [email protected]
Subject: Re: AWS is down.
Hey Patrick,
Please share some info about what enables this, to the extent that you can.
I read in some storage blog that Infinidat for example does 30-40 mics on
average.
With CFs being 2-10 mics, and zHyperLink being ~20 mics, 30-40 for the average
I/O seems pretty good.
The 400-500 mics that you said, is that for a standard IBM/EMC storage box with
zHPF?
If so, with any more tune-ups, or just the above?
- KB
‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Thursday, December 9th, 2021 at 6:47 AM, patrickfalcone7
<[email protected]> wrote:
> Don't like to get into these but I have respect for Seymour and would ask you
> not to fan flames. While I understand Seymour's response he didn't
> necessarily state anything but questioned the possibility.Seymour I did not
> mean to put you in any unnecessary positions and apologize if I did. My post
> was only to state what metrics I've seen in the last couple of years. And
> you'd be surprised at how many might not know that it is possible to get mics
> on average with even backleveled kits.When we did our last array upgrade we
> went from 4 mills to around 400 to 500 mics. To me that is significant and
> ended up being so. I saw workload shifts in time due to the significance of
> the array swap.Sent from my Verizon, Samsung Galaxy smartphone
>
> -------- Original message --------From: Bill Johnson
> [email protected] Date: 12/8/21 4:07 PM
> (GMT-05:00) To: [email protected] Subject: Re: AWS is down. LOLOLOLOL,
> I love when the so called “experts” are proven wrong.Sent from Yahoo Mail for
> iPhoneOn Wednesday, December 8, 2021, 4:04 PM, patrickfalcone7
> [email protected] wrote:Hope you are well. FWIW
> I've seem mics for sometime but mostly under favorable conditions. But lately
> have found mics on average from under 5 on avg. mills.with somewhat newer
> array technology on < z15 CPCs.Sent from my Verizon, Samsung Galaxy
> smartphone-------- Original message --------From: Seymour J Metz
> [email protected] Date: 12/8/21 9:05 AM (GMT-05:00) To: [email protected]
> Subject: Re: AWS is down. On what machine do you complete I/O in a
> microsecond?--Shmuel (Seymour J.)
> Metzhttp://mason.gmu.edu/~smetz3________________________________________From:
> IBM Mainframe Discussion List [[email protected]] on behalf of Bill
> Johnson [[email protected]]Sent: Wednesday,
> December 8, 2021 7:31 AMTo: [email protected]: Re: AWS is
> down.I see someone who has never worked in health care where the mainframe
> processes each drug prescribed and checks for drug interactions in a
> microsecond. Yes, people die if the mainframe isn’t available. It’s also why
> there are plenty of pharmacies open 24 hours and why hospitals have
> pharmacies.Sent from Yahoo Mail for iPhoneOn Wednesday, December 8, 2021,
> 1:33 AM, kekronbekron [email protected]
> wrote:Critical infra in some places, sure, not everyone is denying that.At
> the moment of urgent need, do people really buy something and wait for MF to
> finish processing, for them to be then allowed to continue breathing?What
> happened to the interim stages (logstics etc).It sounds as though failure to
> buy/order something immediately is going to lead to their death... is what's
> being said.Sounds pretty privileged to me.It also sounds like it's assumed
> that mainframes will last 500 years, no?Did the world not exist before
> 1960s?Did people automatically die before 1960s because they didn't have
> MF?Are people and organizations not allowed to be wrong (to their own
> detriment), etc.Are we sure that 100% of all information out there is truth,
> apart from this announcement that AWS wants to replace MF?We can all like MF,
> but don't need to act like it's the sole saviour of humanity.Again that means
> that all technology that evolved since then is a complete waste of time and
> people's efforts.Sounds pretty delusional to me... just to convey that MFs
> are good servers.- KB‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐On Wednesday, December
> 8th, 2021 at 10:00 AM, Bill Johnson
> [email protected] wrote:> AWS had a larger &
> longer outage today, than all the outages on the mainframes I’ve worked on in
> the last 30 years. Much worse. Yet, some here think AWS is going to replace
> the mainframe for critical applications like banking, health care, big
> retail, and insurance. Plus, the mainframe has had plenty of negative posts
> here in the decades I’ve been involved. Nobody dies when Netflix isn’t
> available. (One of the hits from today) people can and do die if the drugs
> they need aren’t available because the computer system is down.>> Sent from
> Yahoo Mail for iPhone>> On Tuesday, December 7, 2021, 11:01 PM, kekronbekron
> [email protected] wrote:>> Not at anybody in
> particular:>> Do we have to make snide remarks because someone else had an
> outage?>> Don't we remember the times when things on the mainframe itself
> went south?>> Has every site's MF env. been 100% available through all these
> decades?>> Is it wrong to fail sometimes?>> Is this the attitude with which
> you'd like to retire, holding onto precious remarks and burning the path
> behind you as you leave?>> What does that say about one's mentality and
> outlook in life?>> - KB>> ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐>> On Wednesday,
> December 8th, 2021 at 3:54 AM, Ronald Wells
> [email protected] wrote:>> > Pretty pic/web
> pages make the $$ did you not get the memo> >> > -----Original Message----->
> >> > From: IBM Mainframe Discussion List [email protected] On Behalf
> Of John McKown> >> > Sent: Tuesday, December 7, 2021 2:57 PM> >> > To:
> [email protected]> >> > Subject: Re: AWS is down.> >> > ** EXTERNAL
> EMAIL - USE CAUTION **> >> > Oh, but they advertise that they can replace
> mainframe. Well, maybe the processing. But not the on site reliability. Of
> course, for WFH, the internet is the week point. And, most likely, also the
> home equipment. I just had to reboot my Windows PC at home because it decided
> to stop talking to my internet router.> >> > On Tue, Dec 7, 2021, 12:49 Bill
> Johnson < [email protected]> wrote:> >> > >
> Amazon Web Services is down. I can’t get to my pictures. Glad we don’t> > >>
> > > have our mainframe running there.> > >> > > For IBM-MAIN subscribe /
> signoff / archive access instructions, send> > >> > > email to
> [email protected] with the message: INFO IBM-MAIN> >> > For IBM-MAIN
> subscribe / signoff / archive access instructions, send email to
> [email protected] with the message: INFO IBM-MAIN> >> > Email
> Disclaimer> >> > This E-mail contains confidential information belonging to
> the sender, which may be legally privileged information. This information is
> intended only for the use of the individual or entity addressed above. If you
> are not the intended recipient, or an employee or agent responsible for
> delivering it to the intended recipient, you are hereby notified that any
> disclosure, copying, distribution, or the taking of any action in reliance on
> the contents of the E-mail or attached files is strictly prohibited.> >> >
> For IBM-MAIN subscribe / signoff / archive access instructions,> >> > send
> email to [email protected] with the message: INFO IBM-MAIN>> For
> IBM-MAIN subscribe / signoff / archive access instructions,>> send email to
> [email protected] with the message: INFO IBM-MAIN>>>
> ----------------------------------------------------------------------------------------------------------------------------------------->>
> For IBM-MAIN subscribe / signoff / archive access instructions,>> send email
> to [email protected] with the message: INFO
> IBM-MAIN----------------------------------------------------------------------For
> IBM-MAIN subscribe / signoff / archive access instructions,send email to
> [email protected] with the message: INFO
> IBM-MAIN----------------------------------------------------------------------For
> IBM-MAIN subscribe / signoff / archive access instructions,send email to
> [email protected] with the message: INFO
> IBM-MAIN----------------------------------------------------------------------For
> IBM-MAIN subscribe / signoff / archive access instructions,send email to
> [email protected] with the message: INFO
> IBM-MAIN----------------------------------------------------------------------For
> IBM-MAIN subscribe / signoff / archive access instructions,send email to
> [email protected] with the message: INFO
> IBM-MAIN----------------------------------------------------------------------For
> IBM-MAIN subscribe / signoff / archive access instructions,send email to
> [email protected] with the message: INFO IBM-MAIN
>
> For IBM-MAIN subscribe / signoff / archive access instructions,
>
> send email to [email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
