bloomin 'eck man! its been a long day - read the newspaper and chill out!!!!!!!!! ;-D
Cheers Peri Stracchino Digital Library Team University of York ext 4082 new email address peri.stracch...@york.ac.uk -----Original Message----- From: Steve Bayliss [mailto:stephen.bayl...@acuityunlimited.net] Sent: 19 May 2011 17:37 To: 'Support and info exchange list for Fedora users.' Subject: Re: [fcrepo-user] REST export API negative array index exception Hi Ben That sounds entirely sensible! Steve > -----Original Message----- > From: Benjamin Armintor [mailto:armin...@gmail.com] > Sent: 19 May 2011 15:04 > To: Support and info exchange list for Fedora users. > Subject: Re: [fcrepo-user] REST export API negative array index exception > > Steve- > Maybe we create an issue for the 3.6 bucket to refactor serializers > to return an inputstream, rather than serializing to a received > inputstream? Seems reasonable enough. > > - Ben > > On 5/19/11, Scott Hammel <sc...@clemson.edu> wrote: > > yeah! i figured Java must have something like that. > > > > have a great day and thanks for all you guys are doing! > > > > Scott > > > > On 05/19/2011 01:10 AM, Stephen Bayliss wrote: > >> Thanks for looking at that Scott. > >> It sounds like we need an alternative such as > >> > http://commons.apache.org/codec/apidocs/org/apache/commons/codec/binary/Ba > se64InputStream.html > >> which > >> does streaming encoding (unlimited size). > >> Steve > >> > >> -----Original Message----- > >> *From:* Scott Hammel [mailto:sc...@clemson.edu] > >> *Sent:* 18 May 2011 21:43 > >> *To:* Support and info exchange list for Fedora users. > >> *Subject:* Re: [fcrepo-user] REST export API negative array index > >> exception > >> > >> One last thing then I'll quit gabbing so this list doesn't stay as > >> busy as the Solr user's list :-) just to summarize: > >> > >> I poked into the source for Apache Commons base64 codec 1.3 at the > >> line indicated in my error logs: it's a line in |encodeBase64() > >> |where a byte array is allocated for storage. To compute the array > >> size to allocate, the method multiplies the size of the incoming > >> binary data array by 8. So 300 MB => 300 * 2^20 * 8 which is > max > >> int, I do believe. > >> > >> Next limit is the practical limit of JVM RAM on a 32-bit server: > >> really slightly less than 2 GB. > >> > >> Next limit is the fact that it looks like a ByteArrayOutputStream > >> uses a byte array as a buffer, and the limit is max int again, but > >> this time for # of bytes in the datastream (appx 2GB). > >> > >> Scott > >> > >> On 05/18/2011 03:53 PM, Scott Hammel wrote: > >>> I see where you are going :-) > >>> > >>> I just ran a 400MB test with an ATOMZip export. Seems to have > worked > >>> just fine. > >>> > >>> A 900MB datastream export to ATOMZip test failed. No exception > >>> generated > >>> in the logs, just an internal server error. I noticed with 3.4.2 > this > >>> can indicate the JVM ran out memory (not surprising if the export > is > >>> still being collected into a ByteArrayOutputStream, I guess). > >>> > >>> Scott > >>> > >>> On 05/18/2011 11:53 AM, Stephen Bayliss wrote: > >>>> Hi Scott > >>>> > >>>> Thanks for that feedback. > >>>> > >>>> It would be interesting to find out if you get the same problem > >>>> using the > >>>> AtomZip export format (info:fedora/fedora-system:ATOMZip-1.1) > >>>> > >>>> Steve > >>>> > >>>>> -----Original Message----- > >>>>> From: Scott Hammel [mailto:sc...@clemson.edu] > >>>>> Sent: 18 May 2011 16:16 > >>>>> To: Support and info exchange list for Fedora users. > >>>>> Subject: Re: [fcrepo-user] REST export API negative array > >>>>> index exception > >>>>> > >>>>> > >>>>> Scott, Steve, > >>>>> > >>>>> REST export in archive format still blows up with Fedora > >>>>> 3.4.2. Actually > >>>>> is crashing on a datastream< 300MB. I gave the JVM 1.5GB of > >>>>> heap, BTW. > >>>>> > >>>>> Regardless, the exception that is in fedora.log is a negative > array > >>>>> index exception. It looks like it is actually occurring down in > the > >>>>> base64 encoder according to the stack trace. > >>>>> > >>>>> It occurs to me that building support for a full archival > >>>>> export of an > >>>>> object in memory for arbitrarily large objects might be > >>>>> pragmatically > >>>>> (is that a word?) impossible: e.g., on 32-bit systems I think > >>>>> you bump > >>>>> into problems giving the JVM more than ~1.8 GB of RAM. That > >>>>> alone limits > >>>>> the size of exportable objects to well under 2GB in that > >>>>> environment. > >>>>> > >>>>> If I was more adept with Java, I'd volunteer to write an > >>>>> exporter that > >>>>> spooled to disk, but alas, I am not and it would take me > >>>>> twice as long > >>>>> as someone who is. :-( > >>>>> > >>>>> I can take one of several alternative paths with my > >>>>> particular project, > >>>>> so it isn't too big an issue to *me* .... I just have to do a > >>>>> little > >>>>> more coding in a middle-tier. Don't know about other folks, of > >>>>> course. > >>>>> > >>>>> -Scott > >>>>> > >>>>> On 05/18/2011 01:08 AM, Stephen Bayliss wrote: > >>>>>> Looking at those lines of code it looks like in theory > >>>>> there would be > >>>>>> a problem there. Once this is confirmed we should probably > >>>>> add a test > >>>>>> case to the large datastreams test suite. And it is likely > >>>>> to cause a > >>>>>> problem with datastreams smaller than 2GB (2^31-1 as maximum > array > >>>>>> index) due to the archive export base64-encoding the content. > >>>>>> > >>>>>>> -----Original Message----- > >>>>>>> From: Scott Prater [mailto:pra...@wisc.edu] > >>>>>>> Sent: 17 May 2011 18:33 > >>>>>>> To: Support and info exchange list for Fedora users. > >>>>>>> Subject: Re: [fcrepo-user] REST export API negative array > index > >>>>>>> exception > >>>>>>> > >>>>>>> > >>>>>>> Yes, trying with the latest stable version (3.4.2) would > >>>>> be useful, > >>>>>>> if you don't mind. There were some lowlevel garbage > collection > >>>>>>> problems that were fixed in the 3.4.2 release; these problems > >>>>>>> manifested themselves in a variety of ways. > >>>>>>> > >>>>>>> I'm not saying this is the issue, but it wouldn't hurt to > >>>>> verify that > >>>>>>> your problem can be reproduced in 3.4.2. > >>>>>>> > >>>>>>> thanks, > >>>>>>> > >>>>>>> -- Scott > >>>>>>> > >>>>>>> On 05/17/2011 12:22 PM, Scott Hammel wrote: > >>>>>>>> I'm pretty sure it is 3.4.0 (from files on the server it > >>>>>>> looks like an > >>>>>>>> August 2010 build. The server is in a totally isolated > >>>>> network with > >>>>>>>> nothing with GUI support that can hit the admin tools). > >>>>>>>> > >>>>>>>> Tomcat is the version bundled with the Fedora installer. > >>>>>>>> > >>>>>>>> Would you like me to be sure I'm running at the latest > >>>>>>> version and try > >>>>>>>> the test scripts again before you go forward? > >>>>>>>> > >>>>>>>> Scott > >>>>>>>> > >>>>>>>> On 05/17/2011 12:45 PM, Scott Prater wrote: > >>>>>>>>> Thanks, Scott. I'll try to reproduce the problem in my > >>>>>>> environment, > >>>>>>>>> Fedora 3.4.2. > >>>>>>>>> > >>>>>>>>> Can you tell me what version of Fedora and Tomcat (or > >>>>> other webapp > >>>>>>>>> server) you're using? > >>>>>>>>> > >>>>>>>>> -- Scott > >>>>>>>>> > >>>>>>>>> On 05/17/2011 11:08 AM, Scott Hammel wrote: > >>>>>>>>>> Hey, Scott, > >>>>>>>>>> > >>>>>>>>>> Thanks for responding. I'm more a C/C++ programmer and > >>>>> not a Java > >>>>>>>>>> programmer (though I sometimes play one on the > >>>>> Internet), so I'm > >>>>>>>>>> just guessing on the array bounds -- feels like something > >>>>>>>>>> incrementing an int into the sign bit, though I'd think > >>>>>>> Java would > >>>>>>>>>> throw some array bounds exception before that happened. > >>>>>>> Figured I'd > >>>>>>>>>> do a little math later maybe to test my hypothesis. > >>>>>>>>>> > >>>>>>>>>> Recall, this was all in a 32-bit environment. I really > >>>>>>> hope it is a > >>>>>>>>>> non-issue and something I'm doing in the end. Note > >>>>>>> disseminating the > >>>>>>>>>> datastream content directly appears to work OK, which > >>>>>>> confuses me a > >>>>>>>>>> little, though I haven't looked to see if the code for > >>>>> that does > >>>>>>>>>> things differently. > >>>>>>>>>> > >>>>>>>>>> Anyway, here's a series of commands (extracted from my > >>>>>>> test scripts) > >>>>>>>>>> that should reproduce the problem: > >>>>>>>>>> > >>>>>>>>>> mkdir /usr/fedora/tomcat/webapps/ROOT/ingestpool > >>>>>>>>>> mkdir /tmp/fedrun > >>>>>>>>>> dir=/tmp/fedrun > >>>>>>>>>> pid=test:pid01 > >>>>>>>>>> > >>>>>>>>>> dd if=/dev/urandom > >>>>>>>>>> of=/usr/fedora/tomcat/webapps/ROOT/ingestpool/sample.bin > bs=1M > >>>>>>>>>> count=400 > >>>>>>>>>> > >>>>>>>>>> ./makefoxml > $pidhttp://localhost:8080/ingestpool/sample.bin> > >>>>>>>>>> $dir/sample.xml > >>>>>>>>>> > >>>>>>>>>> /usr/fedora/client/bin/fedora-ingest.sh f $dir/sample.xml > >>>>>>>>>> info:fedora/fedora-system:FOXML-1.1 localhost:8080 > >>>>>>> fedoraAdmin<insert > >>>>>>>>>> pwd here> http > >>>>>>>>>> > >>>>>>>>>> wget -O $dir/export.xml --auth-no-challenge > >>>>>>> --http-user=fedoraAdmin > >>>>>>>>>> --http-password=<insert pwd here> > >>>>>>>>>> > >>>>>>>>>> > http://localhost:8080/fedora/objects/$pid/export?context=archive > >>>>>>>>>> > >>>>>>>>>> Note: I use the REST call via a wget rather than the > >>>>>>> provided export > >>>>>>>>>> client scripts because it looks to me from the Java heap > >>>>>>> explosion > >>>>>>>>>> that the export scripts must end up doing the export > >>>>> via the SOAP > >>>>>>>>>> API. > >>>>>>>>>> -- > >>>>>>>>>> The content of makefoxml: > >>>>>>>>>> > >>>>>>>>>> #!/bin/bash > >>>>>>>>>> > >>>>>>>>>> #usage: makefoxml<pid> <refurl> > >>>>>>>>>> #escape slashes off the URL > >>>>>>>>>> RF=${2//\//\\/} > >>>>>>>>>> #if you need to escape ampersands as well, uncomment this: > >>>>>>>>>> #RF=${RF//'&'/'\&'} > >>>>>>>>>> > >>>>>>>>>> # make substitutions .... > >>>>>>>>>> sed ' > >>>>>>>>>> s/PID=""/PID="'"$1"'"/ > >>>>>>>>>> s/rdf:about=""/rdf:about="info:fedora\/'"$1"'"/ > >>>>>>>>>> s/dc:identifier>/dc:identifier>'"$1"'/ > >>>>>>>>>> s/REF=""/REF="'"${RF}"'"/ > >>>>>>>>>> '< "foxml_tpl.xml" > >>>>>>>>>> > >>>>>>>>>> -- > >>>>>>>>>> The content of foxml_tmp.xml (the sed script above does > >>>>> the edits > >>>>>>>>>> noted in the xml comments in this template): > >>>>>>>>>> > >>>>>>>>>> <?xml version="1.0" encoding="UTF-8"?> > >>>>>>>>>> <!-- following element: set the PID attribute --> > >>>>>>>>>> <foxml:digitalObject VERSION="1.1" PID="" > >>>>>>>>>> xmlns:foxml="info:fedora/fedora-system:def/foxml#" > >>>>>>>>>> xmlns:xsi="http://www.w3.org/2001/XMLSchema- > instance" > >>>>>>>>>> xsi:schemaLocation="info:fedora/fedora-system:def/foxml# > >>>>>>>>>> http://www.fedora.info/definitions/1/0/foxml1-1.xsd";> > >>>>>>>>>> > >>>>>>>>>> <foxml:objectProperties> > >>>>>>>>>> <foxml:property > >>>>>>>>>> NAME="info:fedora/fedora-system:def/model#state" > >>>>>>>>>> VALUE="A"/> <foxml:property > >>>>>>>>>> NAME="info:fedora/fedora-system:def/model#label" > VALUE=""/> > >>>>>>>>>> <foxml:property > >>>>> NAME="info:fedora/fedora-system:def/model#ownerId" > >>>>>>>>>> VALUE="fedoraAdmin"/> > >>>>>>>>>> </foxml:objectProperties> > >>>>>>>>>> > >>>>>>>>>> <foxml:datastream CONTROL_GROUP="X" ID="RELS-EXT"> > >>>>>>>>>> <foxml:datastreamVersion > >>>>>>>>>> FORMAT_URI="info:fedora/fedora-system:FedoraRELSExt-1.0" > >>>>>>>>>> ID="RELS-EXT.0" LABEL="RDF Statements about > >>>>>>> this Object" > >>>>>>>>>> MIMETYPE="application/rdf+xml"> <foxml:xmlContent> > >>>>>>>>>> <rdf:RDF > >>>>>>>>>> xmlns:dc="http://purl.org/dc/elements/1.1/"; > >>>>>>>>>> > >>>>>>> xmlns:fedora="info:fedora/fedora-system:def/relations- > external#" > >>>>>>> xmlns:fedora-model="info:fedora/fedora-system:def/model#" > >>>>>>> xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/"; > >>>>>>> xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"; > >>>>>>>>>> xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#";> > >>>>>>>>>> <!-- following element: put the PID as the value for > >>>>> the rdf:about > >>>>>>>>>> attribute --> <rdf:description rdf:about=""> > >>>>>>>>>> </rdf:description> > >>>>>>>>>> </rdf:RDF> > >>>>>>>>>> </foxml:xmlContent> > >>>>>>>>>> </foxml:datastreamVersion> > >>>>>>>>>> </foxml:datastream> > >>>>>>>>>> > >>>>>>>>>> <foxml:datastream CONTROL_GROUP="X" ID="DC" STATE="A" > >>>>>>>>>> VERSIONABLE="true"> <foxml:datastreamVersion ID="DC.0" > >>>>>>> LABEL="Dublin > >>>>>>>>>> Core Record" MIMETYPE="text/xml"> <foxml:xmlContent> > >>>>> <oai_dc:dc > >>>>>>>>>> xmlns:dc="http://purl.org/dc/elements/1.1/"; > >>>>>>>>>> > >>>>>>> xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/"; > >>>>>>> xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"; > >>>>>>>>>> > >>>>>>>>>> xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ > >>>>>>>>>> http://www.openarchives.org/OAI/2.0/oai_dc.xsd";> > >>>>>>>>>> <dc:title></dc:title> > >>>>>>>>>> <dc:creator>Test Program</dc:creator> > >>>>>>>>>> <dc:description>A test object</dc:description> > >>>>>>>>>> <!-- following element: put the PID between the tags --> > >>>>>>>>>> <dc:identifier></dc:identifier> </oai_dc:dc> > >>>>>>>>>> </foxml:xmlContent> > >>>>>>>>>> </foxml:datastreamVersion> > >>>>>>>>>> </foxml:datastream> > >>>>>>>>>> > >>>>>>>>>> <foxml:datastream CONTROL_GROUP="M" ID="Content" STATE="A"> > >>>>>>>>>> <foxml:datastreamVersion ID="Content.0" LABEL="This is > >>>>> the object > >>>>>>>>>> content" MIMETYPE=" application/octet-stream"> > >>>>>>>>>> <!-- following element: put the URL to the content file > >>>>>>> as the value > >>>>>>>>>> for the REF attribute --> > >>>>>>>>>> <!-- must be an http URL, e.g., > >>>>>>>>>> http://localhost:8080/ingestpool/foxmldoc.xml --> > >>>>>>>>>> <!-- I just create a directory "ingestpool" under > >>>>>>>>>> /usr/fedora/tomcat/webapps/ROOT and put the files there --> > >>>>>>>>>> <foxml:contentLocation REF="" TYPE="URL" /> > >>>>>>>>>> </foxml:datastreamVersion> </foxml:datastream> > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>> </foxml:digitalObject> > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>> On 05/17/2011 10:00 AM, Scott Prater wrote: > >>>>>>>>>>> Scott, > >>>>>>>>>>> > >>>>>>>>>>> Can you come up with a test case that confirms this > >>>>>>> limitation? If > >>>>>>>>>>> you can provide one, I'll open up a JIRA ticket for the > >>>>>>>>>>> issue. > >>>>>>>>>>> > >>>>>>>>>>> thanks, > >>>>>>>>>>> > >>>>>>>>>>> -- Scott > >>>>>>>>>>> > >>>>>>>>>>> On 05/16/2011 10:45 AM, Scott Hammel wrote: > >>>>>>>>>>>> Oh, I think I see: last line of the serializer's > serialize > >>>>>>>>>>>> function does > >>>>>>>>>>>> this: > >>>>>>>>>>>> bytes.toByteArray() > >>>>>>>>>>>> where bytes is a ByteArrayOutputStream > >>>>>>>>>>>> > >>>>>>>>>>>> I *think* the max size of an array index in Java (32-bit) > is > >>>>>>>>>>>> 2,147,483,647 (i.e., 2^31 - 1, max value of a java > >>>>>>> int). So, this > >>>>>>>>>>>> function will throw an exception if a datastream > >>>>>>> "archive" export > >>>>>>>>>>>> is> ~2 GB. > >>>>>>>>>>>> > >>>>>>>>>>>> scott > >>>>>>>>>>>> > >>>>>>>>>>>> On 05/16/2011 11:00 AM, Scott Hammel wrote: > >>>>>>>>>>>>> Hi, > >>>>>>>>>>>>> > >>>>>>>>>>>>> Running some export tests using Fedora's REST export > >>>>>>> API, I get a > >>>>>>>>>>>>> negative array index Java exception when doing an > >>>>>>> "archive" export of an > >>>>>>>>>>>>> object at around 400 MB (>320 MB,< 450 MB). > >>>>>>>>>>>>> > >>>>>>>>>>>>> Fedora is version 3.4 something; running on 32-bit > >>>>> CentOS 5.5, > >>>>>>>>>>>>> Sun Java 1.6, 21 > >>>>>>>>>>>>> > >>>>>>>>>>>>> Is it just me or has anyone else seen something like > that? > >>>>>>>>>>>>> > >>>>>>>>>>>>> Thanks, > >>>>>>>>>>>>> Scott > >>>>>>>>>>>>> > >>>>>>>>>>>>> > >>>>>>> -------------------------------------------------------------- > --- > >>>>>>>>>>>>> ------------- > >>>>>>>>>>>>> Achieve unprecedented app performance and reliability > What > >>>>>>>>>>>>> every C/C++ and Fortran developer should know. Learn > >>>>> how Intel > >>>>>>>>>>>>> has extended the reach of its > >>>>>>> next-generation tools > >>>>>>>>>>>>> to help boost performance applications - inlcuding > >>>>>>>>>>>>> clusters. > >>>>>>>>>>>>> http://p.sf.net/sfu/intel-dev2devmay > >>>>>>>>>>>>> _______________________________________________ > >>>>>>>>>>>>> Fedora-commons-users mailing list > >>>>>>>>>>>>> Fedora-commons-users@lists.sourceforge.net > >>>>>>>>>>>>> > >>>>>>> https://lists.sourceforge.net/lists/listinfo/fedora-commons- > users > >>>>>>> > >>>>>>> ------------------------------------------------------------------ > >>>>>>>>>>>> ------------ > >>>>>>>>>>>> Achieve unprecedented app performance and reliability > >>>>> What every > >>>>>>>>>>>> C/C++ and Fortran developer should know. Learn how Intel > has > >>>>>>>>>>>> extended the reach of its > >>>>>>> next-generation tools > >>>>>>>>>>>> to help boost performance applications - inlcuding > clusters. > >>>>>>>>>>>> http://p.sf.net/sfu/intel-dev2devmay > >>>>>>>>>>>> _______________________________________________ > >>>>>>>>>>>> Fedora-commons-users mailing list > >>>>>>>>>>>> Fedora-commons-users@lists.sourceforge.net > >>>>>>>>>>>> > >>>>>>> https://lists.sourceforge.net/lists/listinfo/fedora-commons- > users > >>>>> > >>>>> -------------------------------------------------------------------- > >>>>>>>>>> ---------- > >>>>>>>>>> Achieve unprecedented app performance and reliability > >>>>> What every > >>>>>>>>>> C/C++ and Fortran developer should know. Learn how Intel > has > >>>>>>>>>> extended the reach of its > >>>>>>> next-generation tools > >>>>>>>>>> to help boost performance applications - inlcuding > clusters. > >>>>>>>>>> http://p.sf.net/sfu/intel-dev2devmay > >>>>>>>>>> _______________________________________________ > >>>>>>>>>> Fedora-commons-users mailing list > >>>>>>>>>> Fedora-commons-users@lists.sourceforge.net > >>>>>>>>>> > >>>>> https://lists.sourceforge.net/lists/listinfo/fedora-commons- > users > >>>>> > >>>>> -------------------------------------------------------------------- > - > >>>>>>> - > >>>>>>>> -------- > >>>>>>>> Achieve unprecedented app performance and reliability > >>>>>>>> What every C/C++ and Fortran developer should know. > >>>>>>>> Learn how Intel has extended the reach of its > >>>>> next-generation tools > >>>>>>>> to help boost performance applications - inlcuding clusters. > >>>>>>>> http://p.sf.net/sfu/intel-dev2devmay > >>>>>>>> _______________________________________________ > >>>>>>>> Fedora-commons-users mailing list > >>>>>>>> Fedora-commons-users@lists.sourceforge.net > >>>>>>>> > >>>>>>>> https://lists.sourceforge.net/lists/listinfo/fedora-commons-users > >>>>>>> -- > >>>>>>> Scott Prater > >>>>>>> Library, Instructional, and Research Applications (LIRA) > >>>>>>> Division of Information Technology (DoIT) University of > >>>>>>> Wisconsin - madisonpra...@wisc.edu > >>>>>>> > >>>>>>> -------------------------------------------------------------- > >>>>>>> ---------------- > >>>>>>> Achieve unprecedented app performance and reliability > >>>>>>> What every C/C++ and Fortran developer should know. > >>>>>>> Learn how Intel has extended the reach of its > >>>>> next-generation tools > >>>>>>> to help boost performance applications - inlcuding clusters. > >>>>>>> http://p.sf.net/sfu/intel-dev2devmay > >>>>>>> _______________________________________________ > >>>>>>> Fedora-commons-users mailing list > >>>>>>> Fedora-commons-users@lists.sourceforge.net > >>>>>>> https://lists.sourceforge.net/lists/listinfo/fedora-commons- > users > >>>>>>> > >>>>> > >>>>> -------------------------------------------------------------------- > -- > >>>>>> -------- > >>>>>> What Every C/C++ and Fortran developer Should Know! > >>>>>> Read this article and learn how Intel has extended the reach of > >>>>>> its > >>>>>> next-generation tools to help Windows* and Linux* C/C++ and > >>>>>> Fortran > >>>>>> developers boost performance applications - including clusters. > >>>>>> http://p.sf.net/sfu/intel-dev2devmay > >>>>>> _______________________________________________ > >>>>>> Fedora-commons-users mailing list > >>>>>> Fedora-commons-users@lists.sourceforge.net > >>>>>> https://lists.sourceforge.net/lists/listinfo/fedora-commons- > users > >>>>>> > >>>>> -------------------------------------------------------------- > >>>>> ---------------- > >>>>> What Every C/C++ and Fortran developer Should Know! > >>>>> Read this article and learn how Intel has extended the reach of > its > >>>>> next-generation tools to help Windows* and Linux* C/C++ and > Fortran > >>>>> developers boost performance applications - including clusters. > >>>>> http://p.sf.net/sfu/intel-dev2devmay > >>>>> _______________________________________________ > >>>>> Fedora-commons-users mailing list > >>>>> Fedora-commons-users@lists.sourceforge.net > >>>>> https://lists.sourceforge.net/lists/listinfo/fedora-commons- > users > >>>>> > >>>> > >>>> --------------------------------------------------------------------- > --------- > >>>> What Every C/C++ and Fortran developer Should Know! > >>>> Read this article and learn how Intel has extended the reach of > its > >>>> next-generation tools to help Windows* and Linux* C/C++ and > Fortran > >>>> developers boost performance applications - including clusters. > >>>> http://p.sf.net/sfu/intel-dev2devmay > >>>> _______________________________________________ > >>>> Fedora-commons-users mailing list > >>>> Fedora-commons-users@lists.sourceforge.net > >>>> https://lists.sourceforge.net/lists/listinfo/fedora-commons-users > >>>> > >>> > >>> ---------------------------------------------------------------------- > -------- > >>> What Every C/C++ and Fortran developer Should Know! > >>> Read this article and learn how Intel has extended the reach of > its > >>> next-generation tools to help Windows* and Linux* C/C++ and > Fortran > >>> developers boost performance applications - including clusters. > >>> http://p.sf.net/sfu/intel-dev2devmay > >>> _______________________________________________ > >>> Fedora-commons-users mailing list > >>> Fedora-commons-users@lists.sourceforge.net > >>> https://lists.sourceforge.net/lists/listinfo/fedora-commons-users > >>> > >> > >> > >> ----------------------------------------------------------------------- > ------- > >> What Every C/C++ and Fortran developer Should Know! > >> Read this article and learn how Intel has extended the reach of its > >> next-generation tools to help Windows* and Linux* C/C++ and Fortran > >> developers boost performance applications - including clusters. > >> http://p.sf.net/sfu/intel-dev2devmay > >> > >> > >> _______________________________________________ > >> Fedora-commons-users mailing list > >> Fedora-commons-users@lists.sourceforge.net > >> https://lists.sourceforge.net/lists/listinfo/fedora-commons-users > > > > > > -------------------------------------------------------------------------- > ---- > What Every C/C++ and Fortran developer Should Know! > Read this article and learn how Intel has extended the reach of its > next-generation tools to help Windows* and Linux* C/C++ and Fortran > developers boost performance applications - including clusters. > http://p.sf.net/sfu/intel-dev2devmay > _______________________________________________ > Fedora-commons-users mailing list > Fedora-commons-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/fedora-commons-users ---------------------------------------------------------------------------- -- What Every C/C++ and Fortran developer Should Know! Read this article and learn how Intel has extended the reach of its next-generation tools to help Windows* and Linux* C/C++ and Fortran developers boost performance applications - including clusters. http://p.sf.net/sfu/intel-dev2devmay _______________________________________________ Fedora-commons-users mailing list Fedora-commons-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fedora-commons-users ------------------------------------------------------------------------------ What Every C/C++ and Fortran developer Should Know! Read this article and learn how Intel has extended the reach of its next-generation tools to help Windows* and Linux* C/C++ and Fortran developers boost performance applications - including clusters. http://p.sf.net/sfu/intel-dev2devmay _______________________________________________ Fedora-commons-users mailing list Fedora-commons-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fedora-commons-users
