W dniu 2016-07-21 o 16:54, Tim Hare pisze:
[...]
You can test the difference in speed - find a large dataset (or create one),
and FTP it using the OSA network and the other system's non-hipersocket IP
address, then FTP the same file using the hipersocket IP address.
Actually I did it.
I would like to add one other advantage that I took advantage of. When you
define your hipersocket network, you can define a very large MTU size for the
hipersocket network. This can eliminate a lot of segmentation/re-assembly
overhead because the data transmission is broken up into fewer
On Wed, 6 Jul 2016 09:48:37 -0500, Parwez Hamid
wrote:
>The other point to note about OSes when comparing SMC-D with HiperSockets,
>SMC-D is only supported by z/OS whereas HiperSockets is supported by all OSes
>that run on z Systems.
Right. :-(
I am eagerly
Hi,
Would someone be kind to share your Hipersocket Definitions at your
site(Ofcourse masking the critical details).
I am though referring http://www.redbooks.ibm.com/redbooks/pdfs/sg246816.pdf
Just want to get your view on the Dynamic Routing(OMPRCONF) for Hipersocket
so that all the
In addition to the previous comments, there are some potential advantages
to HiperSockets when considering workload portability. You may be able to
ignore some OSA-Express physical and logical configuration considerations
for inter-LPAR communications on, for example, a DR machine. That could
Gil,
I believe it is using the Comms Server APIs.
Ron
-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf
Of Paul Gilmartin
Sent: Wednesday, July 6, 2016 8:44 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: [IBM-MAIN] Real purpose
On Wed, 6 Jul 2016 07:51:24 -0700, Ron Hawkins wrote:
>We're using them for communication between REXX programs on multiple LPARs,
>as well as NJE.
>
>The REXX programs are used for synchronizing test events.
>
Do you use RXSOCKET or a custom interface?
I miss the CMS Pipelines TCPCLIENT stage.
Sent: Wednesday, July 6, 2016 5:24 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: [IBM-MAIN] Real purpose of hipersocket
The more important detractor is that CPU processing to send data via
hipersockets is consumed on CPs, while CPU to send data via OAS cards is
consumed on the OSA card processors
Remember SMC-D is only supported on z13 or z13s at Driver level 27 with z/OS
2.2. With these levels of H/W and OS, when comparing CPU utilization, SMC-D
could be a better option than HiperSockets.
The other point to note about OSes when comparing SMC-D with HiperSockets,
SMC-D is only
On 7/5/2016 9:49 PM, Peter wrote:
I am still trying to understand from the manual what other benefits I can
get from Hipersocket.
It's faster than going out to an OSA, but still an order of magnitude
slower than SMC-D. If you have that on your hardware, use it!
--
Edward E Jaffe
Phoenix
Z/OS 2.2 also provides SMC-D, which I am eager to get a chance to play with.
All the convenience and familiarity of sockets, but with most of the TCP/IP
overhead stripped out (since the underlying transport layer is reliable). It
should have significantly better performance and less CPU
Hipersockets are *MUCH MUCH* faster than exiting one LPAR and returning to
another via OSA cards.
This is done via an internal memory to memory transfer.
All of the OSA overhead as well as the network latency are eliminated.
However, as all LPAR's are in the same SYSPLEX, all of the data should
-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf
Of Peter
Sent: Tuesday, July 5, 2016 11:50 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Real purpose of hipersocket
Cross Posted
Hi
We are planning to implement hipersocket for our sysplex
-MAIN@LISTSERV.UA.EDU] On Behalf
Of Martin Packer
Sent: 06 July, 2016 14:05
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: Real purpose of hipersocket
Free but not consequence-free. So obviously, like its distant cousin IC
links, the CPU to simulate has to come from somewhere. (Recorded in
SMF70PDT
ttps://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2
From: "R.S." <r.skoru...@bremultibank.com.pl>
To: IBM-MAIN@LISTSERV.UA.EDU
Date: 06/07/2016 12:59
Subject:Re: Real purpose of hipersocket
Sent by:IBM Mainframe Discussion List <I
The real purpose of Hipersockets is fast LAN (TCP/IP) communication
between LPARs inside single CPC.
With no real add-on hardware like OSA card.
With no chance for intruders.
With the best performance possible.
It cannot replace CTC. However many CTC exploiters can use TCP/IP links
instead of
You did not mention that the 5 lpars in your sysplex were situated in only 1
machine.
Kees.
-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf
Of Peter
Sent: 06 July, 2016 8:46
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: Real purpose
Parwez
If you have read my post I was very clear that hipersocket are only for
inter LPAR communication.
So apart from above is there any compelling reason to use hipersocket if
the all the LPARS are in same sysplex.
On Jul 6, 2016 12:05 PM, "Parwez Hamid" wrote:
>
HiperSockets is for LPAR to LPAR communications ONLY within the same physical
system. Have you read the following RedBook:
http://www.redbooks.ibm.com/redbooks/pdfs/sg246816.pdf
and also:
https://www.ibm.com/support/knowledgecenter/zosbasics/com.ibm.zos.znetwork/znetwork_85.htm
Cross Posted
Hi
We are planning to implement hipersocket for our sysplex with 5 LPARS in it.
We are already using OSA express card with 10G and our Multi Access Spool
for all the 5 LPARs so we cannot even of implementing hipersocket for NJE.
If my understanding is correct the hipersocket is
20 matches
Mail list logo