Thanks, Ben -- you beat me to it. I added the stack trace to the issue.
-- Scott On 05/19/2011 12:30 PM, Benjamin Armintor wrote: > Tracking this issue here: https://jira.duraspace.org/browse/FCREPO-944 > > On 5/19/11, ps552<peri.stracch...@york.ac.uk> wrote: >> 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 >> > > ------------------------------------------------------------------------------ > 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 -- Scott Prater Library, Instructional, and Research Applications (LIRA) Division of Information Technology (DoIT) University of Wisconsin - Madison pra...@wisc.edu ------------------------------------------------------------------------------ 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
