Re: [CM] Mainframe tech is here to stay: just add innovation

Path	csiph.com!usenet.pasdenom.info!aioe.org!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From	Anne & Lynn Wheeler <lynn@garlic.com>
Newsgroups	comp.misc, alt.folklore.computers
Subject	Re: [CM] Mainframe tech is here to stay: just add innovation
Date	Tue, 24 Jun 2014 13:37:20 -0400
Organization	Wheeler&Wheeler
Lines	78
Message-ID	<m3simuky3z.fsf@garlic.com> (permalink)
References	<locab7$t5a$1@solani.org>
Mime-Version	1.0
Content-Type	text/plain
Injection-Info	mx05.eternal-september.org; posting-host="3f51aa13fbcc0ccfe0b4beace9ec1f30"; logging-data="14145"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX1+QoGIgu7hAzp/SSLWOR0RP4QLLjpqX3TA="
User-Agent	Gnus/5.13 (Gnus v5.13) Emacs/24.3 (gnu/linux)
Cancel-Lock	sha1:fsTEfV2EkgxfK2ZeoKBYEe1yi2I= sha1:rBhauCyafXS1wnkRJC04Yotdyh4=
Xref	csiph.com comp.misc:3821

Cross-posted to 2 groups.

Show key headers only | View raw

there are a couple of such discussions in linkedin mainframe oriented
groups (both ibm employees and customers).

1qtr2014 IBM financials has the equivalent of 17-18 max-configured ec12
(latest ibm mainframe) being sold (56 on annualized basis) @33M which is
rated at 75BIPS ...  or about $440K/BIPS ... is built out of 32nm
technology, six processors in 597mm chip.

a typical cloud megadatacenter has more processing power than the
aggregate of all mainframes in the world today ... a typical component
is e5-2600v1 blade which they build for possibly $1/BIPS. currently
selling are e5-2600v2 and coming on line this year are e5-2600v3 blades
using 14nm technology ... intel has spent $5B on 14nm technology fab and
will feature move from 300mm wafers to 450mm wafers

if the ec12 chip was simply remapped from 32nm to 14nm 450mm wafer ... a
total full years sale of ec12 then is single wafer.

ibm mainframe CKD DASD haven't been manufactured for decades ... being
simulated on industry standard fixed-block disk ... because their
mainstream mainframe operating system doesn't have support for
fixed-block disks.

ibm mainframe i/o "channels" are called "FICON", a heavyweight protocol
layer that runs on industry standard fibre-channel (and drastically
reduces native throughput). A mainframe max I/O benchmark published for
z196 (precursor to ec12) used 104 FICON to get 2M IOPS ... about the
same time a single fibre-channel was announced for e5-2600 claiming over
1M IOPS (two such "native" fibre-channel having higher throughput than
104 FICON).

again for z196 (i don't think they have published for ec12 yet), peak
i/o SSCH operations is 2.2M/sec with all i/o system assist processors
running at 100% busy ... however, they recommend not running the
mainframe SAPs more than 75% busy or 1.5M SSCH/sec.

the published numbers for ec12 claim only 30% increase in I/O over z196
(possibly because of combination of FICON and SAP limitations). max
configured z196 was 80 processors for 50BIPS (625MIPS/processor) while
ec12 is 101 processors for 75BIPS (743MIPS/processor)

posts mentioning FICON
http://www.garlic.com/~lynn/submisc.html#ficon

recent posts mentioning ec12 &/or z196
http://www.garlic.com/~lynn/2014f.html#49 Beyond the EC12
http://www.garlic.com/~lynn/2014f.html#50 Beyond the EC12
http://www.garlic.com/~lynn/2014f.html#51 Beyond the EC12
http://www.garlic.com/~lynn/2014f.html#53 IBM hopes new chip can turn the tables on Intel
http://www.garlic.com/~lynn/2014f.html#65 non-IBM: SONY new tape storage - 185 Terabytes on a tape
http://www.garlic.com/~lynn/2014f.html#67 Is end of mainframe near ?
http://www.garlic.com/~lynn/2014f.html#69 Is end of mainframe near ?
http://www.garlic.com/~lynn/2014f.html#72 Mainframe (in general) running at 100% not always a bad thing
http://www.garlic.com/~lynn/2014f.html#73 Is end of mainframe near ?
http://www.garlic.com/~lynn/2014f.html#75 Is end of mainframe near ?
http://www.garlic.com/~lynn/2014f.html#78 Over in the Mainframe Experts Network LinkedIn group
http://www.garlic.com/~lynn/2014f.html#80 IBM Sales Fall Again, Pressuring Rometty's Profit Goal
http://www.garlic.com/~lynn/2014f.html#86 Is end of mainframe near ?
http://www.garlic.com/~lynn/2014f.html#92 Is end of mainframe near ?
http://www.garlic.com/~lynn/2014g.html#2 Is end of mainframe near ?
http://www.garlic.com/~lynn/2014g.html#4 Is end of mainframe near ?
http://www.garlic.com/~lynn/2014g.html#12 Is end of mainframe near ?
http://www.garlic.com/~lynn/2014g.html#14 Is end of mainframe near ?
http://www.garlic.com/~lynn/2014g.html#20 Is end of mainframe near ?
http://www.garlic.com/~lynn/2014g.html#49 Is end of mainframe near ?
http://www.garlic.com/~lynn/2014g.html#105 Fifty Years of nitpicking definitions, was BASIC,theProgrammingLanguageT
http://www.garlic.com/~lynn/2014h.html#2 Demonstrating Moore's law
http://www.garlic.com/~lynn/2014h.html#4 Demonstrating Moore's law
http://www.garlic.com/~lynn/2014h.html#5 Demonstrating Moore's law
http://www.garlic.com/~lynn/2014h.html#7 Demonstrating Moore's law
http://www.garlic.com/~lynn/2014h.html#8 Demonstrating Moore's law
http://www.garlic.com/~lynn/2014h.html#9 Demonstrating Moore's law
http://www.garlic.com/~lynn/2014h.html#11 Demonstrating Moore's law
http://www.garlic.com/~lynn/2014h.html#41 The Tragedy of Rapid Evolution?

-- 
virtualization experience starting Jan1968, online at home since Mar1970

Thread

[CM] Mainframe tech is here to stay: just add innovation RS Wood  <rsw@therandymon.com> - 2014-06-24 16:51 +0000
  Re: [CM] Mainframe tech is here to stay: just add innovation Anne & Lynn Wheeler <lynn@garlic.com> - 2014-06-24 13:37 -0400
  Re: [CM] Mainframe tech is here to stay: just add innovation Huge <Huge@nowhere.much.invalid> - 2014-06-24 21:24 +0000

csiph-web