Re: [OMPI users] mixed versions of openmpi ? (1.4.1 and 1.4.3)

2010-10-29 Thread Reuti
Am 29.10.2010 um 18:47 schrieb Jeff Squyres:

> On Oct 29, 2010, at 12:40 PM, Reuti wrote:
> 
>>> I'd have to go check 1.4.3 and 1.4.1 to be sure, but I would generally 
>>> *NOT* assume that different versions like this are compatible.
>> 
>> I'm getting confused, as these versions are exactly fitting "x.(y+1).*" 
>> which you mention below. So they should work together by design.
> 
> It depends on what you mean by "work together".
> 
> 1. OMPI provides an ABI guarantee for x.y.* and x.(y+1).*, where y is odd.  
> So if you compile your MPI app with Open MPI v1.4.1, it'll work just fine 
> with 1.4.3.  (the only disclaimer is that this guarantee started with 
> v1.3.2).  Note that y must be odd -- so if you compile your MPI app with 
> v1.4.1, it does *not* necessarily work with v1.5.  Indeed, we broken ABI 
> between the v1.3/v1.4 series and the v1.5 series (our ABI guarantee allows us 
> to do this).

Yep, I read it this way in your first reply.


> 2. OMPI does *not* provide multi-version *interoperability* guarantees.  Say 
> you compile your MPI app against OMPI v1.4.1.  Then you run it across a bunch 
> of nodes, but some nodes have OMPI v1.4.1 on them and others have OMPI v1.4.3 
> (i.e., your app gets libmpi.so from v1.4.1 on some nodes and libmpi.so from 
> v1.4.3 on other nodes).  This is absolutely not guaranteed to work -- we 
> don't even try to maintain this kind of compatibility.

Aha, now I see. When all are the same, it's for sure no problem, but with 
different ones on different nodes you get a mixture of libraries then of course 
for one and the same execution. So, when e.g. the protocol for the message 
which is send to another node changed, it will break.


NB: If I would upgrade my cluster in two steps, I would for a short time adjust 
the queuing system to get nodes for each parallel job where all have the same 
version then.

-- Reuti


> 
> Does that make sense?
> 
> -- 
> Jeff Squyres
> jsquy...@cisco.com
> For corporate legal information go to:
> http://www.cisco.com/web/about/doing_business/legal/cri/
> 
> 
> ___
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users




Re: [OMPI users] mixed versions of openmpi ? (1.4.1 and 1.4.3)

2010-10-29 Thread guillaume ranquet
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I guess we will play it safe and upgrade every cluster at once so that
we won't get bad surprises.

thank you Jeff.
On 10/29/2010 06:40 PM, Reuti wrote:
> Hi,
> 
> Am 29.10.2010 um 18:27 schrieb Jeff Squyres:
> 
>> I'd have to go check 1.4.3 and 1.4.1 to be sure, but I would generally *NOT* 
>> assume that different versions like this are compatible.
> 
> I'm getting confused, as these versions are exactly fitting "x.(y+1).*" which 
> you mention below. So they should work together by design.
> 
> -- Reuti
> 
> 
>> Open MPI makes an ABI promise (that started with version 1.3.2) that all the 
>> releases in a given feature series and its corresponding super-stable series 
>> (i.e., x.y.* and x.(y+1).*, where y is odd) are ABI compatible.  But we make 
>> no guarantees about wire protocols being compatible, or other things like 
>> that.  
>>
>> So in general, it's "pleasantly surprising" if the different releases work 
>> together, but I wouldn't rely on it *at all*.  :-)
>>

If I get it well, ABI compatible means something compiled with x.y.*
will run on x.y+1.* without the need for you to recompile.
mixing x.y and x.y+1 on the same machinefile (and that's what we are
talking about) can only work by accident, not by design.

>>
>> On Oct 29, 2010, at 12:12 PM, guillaume ranquet wrote:
>>
> Hi list,
> I'm sorry to bother you with a stupid question.
> 
> we intend to have for a short period of time, some nodes with 1.4.3 and
> others with 1.4.1 (before upgrading everyone to 1.4.3).
> 
> I made various test and found both versions to be running together quite
> well with a mixed set of nodes.
> 
> my tests were quite simple, I compiled and ran mpi hello_worlds with
> both versions.
> It wouldn't be serious for me to assume both versions fully compatible
> after these tests -and I must admit I lack the time and technical
> knowledge to run further testing.
> 
> has anyone any insight on what have changed that would break compatibility?
> I guess nothing, since they are the same major.minor :)
> 
> 
> regards,
> Guillaume Ranquet.
>>>
___
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
>>
>>
>> -- 
>> Jeff Squyres
>> jsquy...@cisco.com
>> For corporate legal information go to:
>> http://www.cisco.com/web/about/doing_business/legal/cri/
>>
>>
>> ___
>> users mailing list
>> us...@open-mpi.org
>> http://www.open-mpi.org/mailman/listinfo.cgi/users

> ___
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users

-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.16 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJMyvxLAAoJEEzIl7PMEAliTnoH/R4GhehUiFZo6eeSh/Rv9KJc
ZhAJIRTFH0z7+R2V4ggDyIWFVEv0mktQq/WEqQTbGNyVVvhWVFjCxrI7deZ+FkZS
EFv9oIlKM6gNR+cFdoN4xW4ZfiIoCTGddG6XOxLXkZQnhaG30s5UUmIuoBLvgQhb
mTq43WdEPpWsSuyMzo48hizT1PFqpPR101ITnIa2y4T5FC5QktJhbp85HbPaNE2Z
ej7kwXcgLEnTDk9wF4rZRah8vdIdtxwghwGhytVLqMFBCB4MR8hWMYTakJbIOt/7
GkFtOv0D7hruHhl9dNk+o8VyHMQq6bzlqs3UdQxW1Hx1N2w0ngHK6fzfUnYRVVY=
=TsJh
-END PGP SIGNATURE-



smime.p7s
Description: S/MIME Cryptographic Signature


Re: [OMPI users] mixed versions of openmpi ? (1.4.1 and 1.4.3)

2010-10-29 Thread Jeff Squyres
On Oct 29, 2010, at 12:40 PM, Reuti wrote:

>> I'd have to go check 1.4.3 and 1.4.1 to be sure, but I would generally *NOT* 
>> assume that different versions like this are compatible.
> 
> I'm getting confused, as these versions are exactly fitting "x.(y+1).*" which 
> you mention below. So they should work together by design.

It depends on what you mean by "work together".

1. OMPI provides an ABI guarantee for x.y.* and x.(y+1).*, where y is odd.  So 
if you compile your MPI app with Open MPI v1.4.1, it'll work just fine with 
1.4.3.  (the only disclaimer is that this guarantee started with v1.3.2).  Note 
that y must be odd -- so if you compile your MPI app with v1.4.1, it does *not* 
necessarily work with v1.5.  Indeed, we broken ABI between the v1.3/v1.4 series 
and the v1.5 series (our ABI guarantee allows us to do this).

2. OMPI does *not* provide multi-version *interoperability* guarantees.  Say 
you compile your MPI app against OMPI v1.4.1.  Then you run it across a bunch 
of nodes, but some nodes have OMPI v1.4.1 on them and others have OMPI v1.4.3 
(i.e., your app gets libmpi.so from v1.4.1 on some nodes and libmpi.so from 
v1.4.3 on other nodes).  This is absolutely not guaranteed to work -- we don't 
even try to maintain this kind of compatibility.

Does that make sense?

-- 
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/




Re: [OMPI users] mixed versions of openmpi ? (1.4.1 and 1.4.3)

2010-10-29 Thread Reuti
Hi,

Am 29.10.2010 um 18:27 schrieb Jeff Squyres:

> I'd have to go check 1.4.3 and 1.4.1 to be sure, but I would generally *NOT* 
> assume that different versions like this are compatible.

I'm getting confused, as these versions are exactly fitting "x.(y+1).*" which 
you mention below. So they should work together by design.

-- Reuti


> Open MPI makes an ABI promise (that started with version 1.3.2) that all the 
> releases in a given feature series and its corresponding super-stable series 
> (i.e., x.y.* and x.(y+1).*, where y is odd) are ABI compatible.  But we make 
> no guarantees about wire protocols being compatible, or other things like 
> that.  
> 
> So in general, it's "pleasantly surprising" if the different releases work 
> together, but I wouldn't rely on it *at all*.  :-)
> 
> 
> On Oct 29, 2010, at 12:12 PM, guillaume ranquet wrote:
> 
>> -BEGIN PGP SIGNED MESSAGE-
>> Hash: SHA1
>> 
>> Hi list,
>> I'm sorry to bother you with a stupid question.
>> 
>> we intend to have for a short period of time, some nodes with 1.4.3 and
>> others with 1.4.1 (before upgrading everyone to 1.4.3).
>> 
>> I made various test and found both versions to be running together quite
>> well with a mixed set of nodes.
>> 
>> my tests were quite simple, I compiled and ran mpi hello_worlds with
>> both versions.
>> It wouldn't be serious for me to assume both versions fully compatible
>> after these tests -and I must admit I lack the time and technical
>> knowledge to run further testing.
>> 
>> has anyone any insight on what have changed that would break compatibility?
>> I guess nothing, since they are the same major.minor :)
>> 
>> 
>> regards,
>> Guillaume Ranquet.
>> -BEGIN PGP SIGNATURE-
>> Version: GnuPG v2.0.16 (GNU/Linux)
>> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
>> 
>> iQEcBAEBAgAGBQJMyvJTAAoJEEzIl7PMEAliT4kH/RY4WXhEO5R8H3DNIWW7Y91z
>> 6q4BrymLrBSl7rnnEgALBMiPGK9lQgqEtv9k3xCFbfOfWXIFreIHH8ZFDzt1BjQI
>> TZ58SwVE9CIMmESoJ1P52R+WCbKYur3U2eda//1cfnZ28ZYjnKN/xYlT/wv8hqg3
>> GsW+seMR8X+1nNFkH1UQHIBVO2cXaK24BtSe4cvDFaMaUbe0Qlmxg55BbCSYB4ED
>> VBbplp/ty0tojmZdJLqSsp7nZ84oCfvAfZf16fJTDHNYhUSvNz/fldnxWrm7WTUb
>> VzM94yJf0IHfNAB/YvpXECGFL9cPWeG/F6Bm+r6GSMRvd0MeLbp1HWJTbVYlCwo=
>> =NwEP
>> -END PGP SIGNATURE-
>> 
>> ___
>> users mailing list
>> us...@open-mpi.org
>> http://www.open-mpi.org/mailman/listinfo.cgi/users
> 
> 
> -- 
> Jeff Squyres
> jsquy...@cisco.com
> For corporate legal information go to:
> http://www.cisco.com/web/about/doing_business/legal/cri/
> 
> 
> ___
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users




Re: [OMPI users] mixed versions of openmpi ? (1.4.1 and 1.4.3)

2010-10-29 Thread Jeff Squyres
I'd have to go check 1.4.3 and 1.4.1 to be sure, but I would generally *NOT* 
assume that different versions like this are compatible.

Open MPI makes an ABI promise (that started with version 1.3.2) that all the 
releases in a given feature series and its corresponding super-stable series 
(i.e., x.y.* and x.(y+1).*, where y is odd) are ABI compatible.  But we make no 
guarantees about wire protocols being compatible, or other things like that.  

So in general, it's "pleasantly surprising" if the different releases work 
together, but I wouldn't rely on it *at all*.  :-)


On Oct 29, 2010, at 12:12 PM, guillaume ranquet wrote:

> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
> 
> Hi list,
> I'm sorry to bother you with a stupid question.
> 
> we intend to have for a short period of time, some nodes with 1.4.3 and
> others with 1.4.1 (before upgrading everyone to 1.4.3).
> 
> I made various test and found both versions to be running together quite
> well with a mixed set of nodes.
> 
> my tests were quite simple, I compiled and ran mpi hello_worlds with
> both versions.
> It wouldn't be serious for me to assume both versions fully compatible
> after these tests -and I must admit I lack the time and technical
> knowledge to run further testing.
> 
> has anyone any insight on what have changed that would break compatibility?
> I guess nothing, since they are the same major.minor :)
> 
> 
> regards,
> Guillaume Ranquet.
> -BEGIN PGP SIGNATURE-
> Version: GnuPG v2.0.16 (GNU/Linux)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
> 
> iQEcBAEBAgAGBQJMyvJTAAoJEEzIl7PMEAliT4kH/RY4WXhEO5R8H3DNIWW7Y91z
> 6q4BrymLrBSl7rnnEgALBMiPGK9lQgqEtv9k3xCFbfOfWXIFreIHH8ZFDzt1BjQI
> TZ58SwVE9CIMmESoJ1P52R+WCbKYur3U2eda//1cfnZ28ZYjnKN/xYlT/wv8hqg3
> GsW+seMR8X+1nNFkH1UQHIBVO2cXaK24BtSe4cvDFaMaUbe0Qlmxg55BbCSYB4ED
> VBbplp/ty0tojmZdJLqSsp7nZ84oCfvAfZf16fJTDHNYhUSvNz/fldnxWrm7WTUb
> VzM94yJf0IHfNAB/YvpXECGFL9cPWeG/F6Bm+r6GSMRvd0MeLbp1HWJTbVYlCwo=
> =NwEP
> -END PGP SIGNATURE-
> 
> ___
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users


-- 
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/