September 2011 - Mpich2-maint - mailman3.cels.anl.gov

[fortran error]
by Dinesh Prasanth 29 Sep '11

29 Sep '11

I am trying to install mpi.. Fortran is not installed.. I dont find apt package that is to be installed.. please help me with ow to proceed with installing f77 and g77 and proceed with the other installation procedure.. Thanks in advance.. Dinesh Prasanth

1 0

Re: [mpich2-maint] #1532: xhpl compiled with mpich2 has a memory leak
by Guillaume Mercier 29 Sep '11

29 Sep '11

Hello, My understanding is that you have two issues: 1- Your application takes a long time to start with MPICH2:Nemesis:MX 2- There is a memory leak that prevents you from using your application. Both may be related. Actually, the initialization phase in the MX Nemesis module is not optimized, meaning that we create all connections at startup. Currently, a scheme based on dynamic connections is not implemented which would speed-up nicely the init phase. Unfortunately, I can't do this now, and probably won't in the near future. Now, I'm not aware of any memory leak, but there can be one and I can investigate in order to fix this issue. Would that help? Regards, Guillaume Le 09/28/2011 09:37 PM, Randall Martin a ï¿½crit : > Yes we have myricom's version, but it is very slow (unusable). Your > version of mpich2 is much faster, but this memory leak prevents us from > being able to use it with xhpl. It does work fine on a small number of > nodes with xhpl. The leak manifests itself when you begin growing the > number of cores. Watching on a node with top shows that 2 of the 8 xhpl > processes on a node seem to grow without bound until all memory is > exhausted and xhpl dies. > > > On 9/28/11 3:31 PM, "mpich2"<mpich2-maint(a)mcs.anl.gov> wrote: > >> #1532: xhpl compiled with mpich2 has a memory leak >> ------------------------------+------------------------------------------- >> -- >> Reporter: wolf@ï¿½ | Type: bug >> Status: new | Priority: major >> Milestone: | Component: mpich2 >> Keywords: | >> ------------------------------+------------------------------------------- >> -- >> >> Comment(by thakur): >> >> Have you tried using the MPICH2-MX available from Myricom? >> http://www.myricom.com/kb/index.php?title=MPI_Implementations_with_MX >> >> -- >> Ticket URL: >> <http://trac.mcs.anl.gov/projects/mpich2/ticket/1532#comment:1> >> mpich2<http://www.mcs.anl.gov/research/projects/mpich2> >> MPICH2 is a high-performance and widely portable implementation of MPI > _______________________________________________ > mpich2-maint mailing list > mpich2-maint(a)lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/mpich2-maint

1 0

Re: [mpich2-maint] #1533: xhpl compiled with mpich2 taking insanely long time to start on 14000+ cores
by Pavan Balaji 28 Sep '11

28 Sep '11

Please use mpich-discuss for comments. Emails to mpich2-maint are not archived and lost. On 09/28/2011 03:16 PM, Randall Martin wrote: > I'm to sure if other programs have similar behavior, and unfortunately I > won't able to try again on our entire cluster until our next benchmark > outage next April. I do know that USC had the exact same startup issue on > their cluster benchmark and had to abandon mpich2 as well. > > We found this link that talks about issues with Hydra that are disturbing > and may be related: > > http://wiki.mcs.anl.gov/mpich2/index.php/Hydra_Process_Management_Framework > #Ongoing_Work > > "Pre-launched and Pre-connected Proxies: One of the issues on process > management on large scale systems is the time taken to launch processes as > well as initiating their interaction. For example, in the case of MPICH2, > even after the application processes have been launched, each process does > not know anything about the other processes. Thus, it has to interact with > the process manager which exchanges this information. This is typically at > start time, so is not noticed in benchmark performance numbers, but as > system sizes grow larger, this can be a major concern. In our pre-launched > and pre-connected proxy model, we allow the proxies to be launched by a > system administrator (or the user himself) when at "initialization time", > i.e., one initialization is sufficient to run any number of jobs. Once > launched, the proxies connect with each other. For scalable network stacks > such as Myrinet MX and InfiniBand UD, the amount of resources used does > not increase largely with the system size, so an all-to-all connection is > preferred. Once connected, each proxy can interact with any other proxy to > exchange information with a single hop, thus allowing the applications to > be not affected by system scale in a lot of aspects." > > > > -Randy > > > On 9/28/11 3:50 PM, "mpich2"<mpich2-maint(a)mcs.anl.gov> wrote: > >> #1533: xhpl compiled with mpich2 taking insanely long time to start on >> 14000+ >> cores >> ------------------------------+------------------------------------------- >> -- >> Reporter: wolf@ï¿½ | Type: bug >> Status: new | Priority: major >> Milestone: | Component: mpich2 >> Keywords: | >> ------------------------------+------------------------------------------- >> -- >> >> Comment(by thakur): >> >> Is it just xhpl or any program? Can you try running the cpi example from >> the examples directory. >> >> -- >> Ticket URL: >> <http://trac.mcs.anl.gov/projects/mpich2/ticket/1533#comment:1> >> mpich2<http://www.mcs.anl.gov/research/projects/mpich2> >> MPICH2 is a high-performance and widely portable implementation of MPI > > _______________________________________________ > mpich2-maint mailing list > mpich2-maint(a)lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/mpich2-maint -- Pavan Balaji http://www.mcs.anl.gov/~balaji

1 0

Re: [mpich2-maint] #1533: xhpl compiled with mpich2 taking insanely long time to start on 14000+ cores
by Randall Martin 28 Sep '11

28 Sep '11

I'm to sure if other programs have similar behavior, and unfortunately I won't able to try again on our entire cluster until our next benchmark outage next April. I do know that USC had the exact same startup issue on their cluster benchmark and had to abandon mpich2 as well. We found this link that talks about issues with Hydra that are disturbing and may be related: http://wiki.mcs.anl.gov/mpich2/index.php/Hydra_Process_Management_Framework #Ongoing_Work "Pre-launched and Pre-connected Proxies: One of the issues on process management on large scale systems is the time taken to launch processes as well as initiating their interaction. For example, in the case of MPICH2, even after the application processes have been launched, each process does not know anything about the other processes. Thus, it has to interact with the process manager which exchanges this information. This is typically at start time, so is not noticed in benchmark performance numbers, but as system sizes grow larger, this can be a major concern. In our pre-launched and pre-connected proxy model, we allow the proxies to be launched by a system administrator (or the user himself) when at "initialization time", i.e., one initialization is sufficient to run any number of jobs. Once launched, the proxies connect with each other. For scalable network stacks such as Myrinet MX and InfiniBand UD, the amount of resources used does not increase largely with the system size, so an all-to-all connection is preferred. Once connected, each proxy can interact with any other proxy to exchange information with a single hop, thus allowing the applications to be not affected by system scale in a lot of aspects." -Randy On 9/28/11 3:50 PM, "mpich2" <mpich2-maint(a)mcs.anl.gov> wrote: >#1533: xhpl compiled with mpich2 taking insanely long time to start on >14000+ >cores >------------------------------+------------------------------------------- >-- > Reporter: wolf@Š | Type: bug > Status: new | Priority: major >Milestone: | Component: mpich2 > Keywords: | >------------------------------+------------------------------------------- >-- > >Comment(by thakur): > > Is it just xhpl or any program? Can you try running the cpi example from > the examples directory. > >-- >Ticket URL: ><http://trac.mcs.anl.gov/projects/mpich2/ticket/1533#comment:1> >mpich2 <http://www.mcs.anl.gov/research/projects/mpich2> >MPICH2 is a high-performance and widely portable implementation of MPI

1 0

Re: [mpich2-maint] #1532: xhpl compiled with mpich2 has a memory leak
by Randall Martin 28 Sep '11

28 Sep '11

Myricom's mpich1 performs much better than their mpich2, but their mpich1 is not as fast as your mpich2 in our small tests. I am not sure if the leak exists using TCP over myrinet, but I can test that easily. I'll test w/tcp over myrinet and get back to you with the results (may take a few days depending on how busy our cluster is). -Randy On 9/28/11 3:42 PM, "Rajeev Thakur" <thakur(a)mcs.anl.gov> wrote: >I am surprised that Myricom's version would be that slow. > >Does the memory leak also happen with the default TCP device in MPICH2 >(TCP over Myrinet). > >Rajeev > > >On Sep 28, 2011, at 2:37 PM, Randall Martin wrote: > >> Yes we have myricom's version, but it is very slow (unusable). Your >> version of mpich2 is much faster, but this memory leak prevents us from >> being able to use it with xhpl. It does work fine on a small number of >> nodes with xhpl. The leak manifests itself when you begin growing the >> number of cores. Watching on a node with top shows that 2 of the 8 xhpl >> processes on a node seem to grow without bound until all memory is >> exhausted and xhpl dies. >> >> >> On 9/28/11 3:31 PM, "mpich2" <mpich2-maint(a)mcs.anl.gov> wrote: >> >>> #1532: xhpl compiled with mpich2 has a memory leak >>> >>>------------------------------+----------------------------------------- >>>-- >>> -- >>> Reporter: wolf@Š | Type: bug >>> Status: new | Priority: major >>> Milestone: | Component: mpich2 >>> Keywords: | >>> >>>------------------------------+----------------------------------------- >>>-- >>> -- >>> >>> Comment(by thakur): >>> >>> Have you tried using the MPICH2-MX available from Myricom? >>> http://www.myricom.com/kb/index.php?title=MPI_Implementations_with_MX >>> >>> -- >>> Ticket URL: >>> <http://trac.mcs.anl.gov/projects/mpich2/ticket/1532#comment:1> >>> mpich2 <http://www.mcs.anl.gov/research/projects/mpich2> >>> MPICH2 is a high-performance and widely portable implementation of MPI >> >> _______________________________________________ >> mpich2-maint mailing list >> mpich2-maint(a)lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/mpich2-maint >

1 0

Re: [mpich2-maint] #1532: xhpl compiled with mpich2 has a memory leak
by Rajeev Thakur 28 Sep '11

28 Sep '11

I am surprised that Myricom's version would be that slow. Does the memory leak also happen with the default TCP device in MPICH2 (TCP over Myrinet). Rajeev On Sep 28, 2011, at 2:37 PM, Randall Martin wrote: > Yes we have myricom's version, but it is very slow (unusable). Your > version of mpich2 is much faster, but this memory leak prevents us from > being able to use it with xhpl. It does work fine on a small number of > nodes with xhpl. The leak manifests itself when you begin growing the > number of cores. Watching on a node with top shows that 2 of the 8 xhpl > processes on a node seem to grow without bound until all memory is > exhausted and xhpl dies. > > > On 9/28/11 3:31 PM, "mpich2" <mpich2-maint(a)mcs.anl.gov> wrote: > >> #1532: xhpl compiled with mpich2 has a memory leak >> ------------------------------+------------------------------------------- >> -- >> Reporter: wolf@Š | Type: bug >> Status: new | Priority: major >> Milestone: | Component: mpich2 >> Keywords: | >> ------------------------------+------------------------------------------- >> -- >> >> Comment(by thakur): >> >> Have you tried using the MPICH2-MX available from Myricom? >> http://www.myricom.com/kb/index.php?title=MPI_Implementations_with_MX >> >> -- >> Ticket URL: >> <http://trac.mcs.anl.gov/projects/mpich2/ticket/1532#comment:1> >> mpich2 <http://www.mcs.anl.gov/research/projects/mpich2> >> MPICH2 is a high-performance and widely portable implementation of MPI > > _______________________________________________ > mpich2-maint mailing list > mpich2-maint(a)lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/mpich2-maint

1 0

Re: [mpich2-maint] #1532: xhpl compiled with mpich2 has a memory leak
by Randall Martin 28 Sep '11

28 Sep '11

Yes we have myricom's version, but it is very slow (unusable). Your version of mpich2 is much faster, but this memory leak prevents us from being able to use it with xhpl. It does work fine on a small number of nodes with xhpl. The leak manifests itself when you begin growing the number of cores. Watching on a node with top shows that 2 of the 8 xhpl processes on a node seem to grow without bound until all memory is exhausted and xhpl dies. On 9/28/11 3:31 PM, "mpich2" <mpich2-maint(a)mcs.anl.gov> wrote: >#1532: xhpl compiled with mpich2 has a memory leak >------------------------------+------------------------------------------- >-- > Reporter: wolf@Š | Type: bug > Status: new | Priority: major >Milestone: | Component: mpich2 > Keywords: | >------------------------------+------------------------------------------- >-- > >Comment(by thakur): > > Have you tried using the MPICH2-MX available from Myricom? > http://www.myricom.com/kb/index.php?title=MPI_Implementations_with_MX > >-- >Ticket URL: ><http://trac.mcs.anl.gov/projects/mpich2/ticket/1532#comment:1> >mpich2 <http://www.mcs.anl.gov/research/projects/mpich2> >MPICH2 is a high-performance and widely portable implementation of MPI

1 0

ICNC'12-FSKD'12 Submission Deadline 15 November: IEEE Xplore/EI Compendex/ISI
by ICNC FSKD 26 Sep '11

26 Sep '11

Dear Colleague, We cordially invite you to submit a paper or invited session proposal to the upcoming 8th International Conference on Natural Computation (ICNC'12) and the 9th International Conference on Fuzzy Systems and Knowledge Discovery (FSKD'12), to be jointly held from 29-31 May 2012, in Chongqing, China. Renowned as the Mountain City, Chongqing is a magnet for visitors from home and abroad for its cultural heritage and numerous attractions. There are many karst caves, hot springs, and gorges in the area. Major tourist spots in and near Chongqing include Dazu Grottoes (rock carvings began in the Tang Dynasty 650 A.D.), Three Gorges, Jinyun Mountain Natural Reserve, Hongya Cave, Shibaozhai, Wulong Karst, etc.. All papers in the conference proceedings will be included in IEEE Xplore and indexed by both EI Compendex and ISTP as with the past ICNC-FSKD conferences. Extended versions of selected best papers will appear in an ICNC-FSKD special issue of an SCI-indexed journal (ICNC’12: Conference Record Number 19846; IEEE Catalog Number CFP12CNC-ART; ISBN 978-1-4577-2133-5; FSKD’12: Conference Record Number 19847; IEEE Catalog Number CFP12FSK-ART; ISBN 978-1-4673-0024-7). ICNC'12-FSKD'12 is technically co-sponsored by the IEEE Circuits and Systems Society. ICNC-FSKD is a premier international forum for scientists and researchers to present the state of the art of data mining and intelligent methods inspired from nature, particularly biological, linguistic, and physical systems, with applications to computers, circuits, systems, control, communications, and more. This is an exciting and emerging interdisciplinary area in which a wide range of theory and methodologies are being investigated and developed to tackle complex and challenging problems. The registration fee of US-D400 includes proceedings, lunches, dinners, banquet, coffee breaks, and all technical sessions. To promote international participation of researchers from outside the country/region where the conference is held (i.e., China’s mainland), researchers outside of China’s mainland are encouraged to propose invited sessions. An honorarium of US-D400 will be enjoyed by the organizer(s) for each completed (with at least 6 registered papers) invited session. The first author of each paper in an invited session must not be affiliated with an organization in China’s mainland. "(Invited Paper)" may be added below the title of each paper in the invited sessions. Invited session organizers will solicit submissions, conduct reviews and recommend accept/reject decisions on the submitted papers. Invited session organizers will be able to set their own submission and review schedules, as long as a set of recommended papers is determined by 21 March 2012. Each invited session proposal should include: (1) the name, bio, and contact information of each organizer of the invited session; (2) the title and a short synopsis of the invited session. Please send your proposal to icnc-fskd(a)cqupt.edu.cn For more information, visit the conference web page: http://icnc-fskd.cqupt.edu.cn If you have any questions after visiting the conference web page, please email the secretariat at icnc-fskd(a)cqupt.edu.cn Join us at this major event in beautiful Chongqing !!! Organizing Committee icnc-fskd(a)cqupt.edu.cn P.S.: Kindly forward to your colleagues and students in your school/department. If you wish to unsubscribe, in which case we apologize, please reply with " unsubscribe mpi-maint(a)mcs.anl.gov " in your email subject. Thanks.

1 0

ICNC'12-FSKD'12 Submission Deadline 15 November: IEEE Xplore/EI Compendex/ISI
by ICNC FSKD 26 Sep '11

26 Sep '11

Dear Colleague, We cordially invite you to submit a paper or invited session proposal to the upcoming 8th International Conference on Natural Computation (ICNC'12) and the 9th International Conference on Fuzzy Systems and Knowledge Discovery (FSKD'12), to be jointly held from 29-31 May 2012, in Chongqing, China. Renowned as the Mountain City, Chongqing is a magnet for visitors from home and abroad for its cultural heritage and numerous attractions. There are many karst caves, hot springs, and gorges in the area. Major tourist spots in and near Chongqing include Dazu Grottoes (rock carvings began in the Tang Dynasty 650 A.D.), Three Gorges, Jinyun Mountain Natural Reserve, Hongya Cave, Shibaozhai, Wulong Karst, etc.. All papers in the conference proceedings will be included in IEEE Xplore and indexed by both EI Compendex and ISTP as with the past ICNC-FSKD conferences. Extended versions of selected best papers will appear in an ICNC-FSKD special issue of an SCI-indexed journal (ICNC’12: Conference Record Number 19846; IEEE Catalog Number CFP12CNC-ART; ISBN 978-1-4577-2133-5; FSKD’12: Conference Record Number 19847; IEEE Catalog Number CFP12FSK-ART; ISBN 978-1-4673-0024-7). ICNC'12-FSKD'12 is technically co-sponsored by the IEEE Circuits and Systems Society. ICNC-FSKD is a premier international forum for scientists and researchers to present the state of the art of data mining and intelligent methods inspired from nature, particularly biological, linguistic, and physical systems, with applications to computers, circuits, systems, control, communications, and more. This is an exciting and emerging interdisciplinary area in which a wide range of theory and methodologies are being investigated and developed to tackle complex and challenging problems. The registration fee of US-D400 includes proceedings, lunches, dinners, banquet, coffee breaks, and all technical sessions. To promote international participation of researchers from outside the country/region where the conference is held (i.e., China’s mainland), researchers outside of China’s mainland are encouraged to propose invited sessions. An honorarium of US-D400 will be enjoyed by the organizer(s) for each completed (with at least 6 registered papers) invited session. The first author of each paper in an invited session must not be affiliated with an organization in China’s mainland. "(Invited Paper)" may be added below the title of each paper in the invited sessions. Invited session organizers will solicit submissions, conduct reviews and recommend accept/reject decisions on the submitted papers. Invited session organizers will be able to set their own submission and review schedules, as long as a set of recommended papers is determined by 21 March 2012. Each invited session proposal should include: (1) the name, bio, and contact information of each organizer of the invited session; (2) the title and a short synopsis of the invited session. Please send your proposal to icnc-fskd(a)cqupt.edu.cn For more information, visit the conference web page: http://icnc-fskd.cqupt.edu.cn If you have any questions after visiting the conference web page, please email the secretariat at icnc-fskd(a)cqupt.edu.cn Join us at this major event in beautiful Chongqing !!! Organizing Committee icnc-fskd(a)cqupt.edu.cn P.S.: Kindly forward to your colleagues and students in your school/department. If you wish to unsubscribe, in which case we apologize, please reply with " unsubscribe mpich2-maint(a)mcs.anl.gov " in your email subject. Thanks.

1 0

至于您信不信，反正我是信了
by CFP 22 Sep '11

22 Sep '11

至于您信不信，反正我是信了鑫宇翻译专业诚信超值淘宝首家会议超市 http://xinyustudio.taobao.com 如果您正在写英文科技论文您的英语很牛，时间又很多，请忽略该网页！你是不是每年都在为缺三大检索论文担心凭职称而发愁？你是不是经常碰到想参考一下期刊网某个论文，却因为不是教育网而无法下载而郁闷？你是不是经常为论文的排版画图不熟练耗费太多半天而心烦？你是不是引用别人的论文，又顾虑将来被抓住把柄而担忧？你是不是等论文赶紧检索好毕业？你是不是最近正在找容易中且影响因子高的SCI、EI、ISTP期刊和会议？你是不是想快点知道去年发表的论文到底检索了没有？你是不是想找快速渠道搞定论文投稿和检索就好了，一篇SCI、EI期刊检索一般要一年，时间太长，让人等得受不了！有木有合适方式的快速投中EI检索论文？！有木有??? 有！！！鑫宇检测，翻译，排版，提交，注册，检索报告总有一样适合您鑫宇工作室为您解忧，只要您给我们中文，您的论文就能EI检索。鑫宇，淘宝首家：EI & ISTP论文一站式、订单式、一条龙解决方案！什么是一站式？或许您不相信，但您可以自己验证，鑫宇很可能是淘宝首家，目前唯一以自有翻译工作室为核心为客户提供EI&ISTP检索论文从中文->检测->翻译->排版->提交->注册->检索->证明的一条龙服务解决方案的商家。什么是订单式？鑫宇充分考虑客户需求，从来不强制用户消费，鑫宇既可以提供订单定制式套餐服务，亦可以提供单项服务，按项收费，最低1元起。如果您月薪3000以上，又没太多时间，想尽快发表英文EI检索论文只要花5分钟看完以下内容，您就知道鑫宇到底能够为您做什么？！解决什么问题？ ①鑫宇会议怎么样找全方位的最新的、最符合主题的、100%检索的国际会议信息？教育网的中国学术会议在线网站每年都有成千上万的会议信息，各种国际学术会议层出不穷，有的滥竽充数，有的挂羊头卖狗肉，整天垃圾邮件一堆，您是不是也深受其害。很多人无法甄别国际会议真假和可靠性。我们有专职客服免费为您推介，帮您火眼金睛甄别垃圾会议，借助我们的快速通道确保您的论文被符合主题的检索会议快速录用。找鑫宇，帮推荐。省心客服QQ：100576934 ②鑫宇翻译翻译能不能不要这么让我头疼？留学过的大牛大都直接用英文写顶级论文，在SCI发表，我们必须得承认，这样的牛人不在我们的服务对象之列。绝大部分国内的科研学术工作者都不曾有英语国家留学的经历，写论文往往从中文开始，一篇5000字符的中文论文，写好核心学术内容已经相当费脑力和体力，想在EI期刊或会议上发表，还得字斟句酌翻译，想想都头大。为了保证质量，为了省钱，没办法，也有作者愿意克服重重困难，按英语专八译员每天人工翻译2000字符的效率自己翻译论文，大不了花费3天时间。很多作者可能忘了一点，没有计算过，如果自己每月工资3000以上，每天工作8小时，以20个工作日计算时薪约20元，翻译3天的时间成本转化为金额的话，至少也是500元。没完，翻译OK了，还得画图表，排版，提交，注册再怎么也得1-2天，这个时间成本又要加上300元左右。如此算来，大部分月薪3000元左右的作者，自己搞掂论文大概要花至少800元的时间成本，这还是按照专八翻译效率。不过月薪3000元左右的人，很少计算自己劳动的时间成本，都愿意自己翻译，自己排版，自己提交，自己注册，自己搞检索报告，在他们看来，凡事自己能做的事情都尽量自己做就是在替自己省钱！！省时间！！！殊不知，与其在自己不擅长的领域耗费时间做无专业价值的重复劳动，不如把这几天时间用在查阅更多文献，撰写更多论文，做更多的时间，或许会产生更大的专业价值！！月薪3000元左右的人，很少计算自己劳动的时间成本，他们看到的就是如何更好地省钱，而不是省时间去提升自己的时间价值，专业价值，人生价值。所以他们的月薪总是维持在3000左右，因为他们没有时间去计算并提升自己的劳动价值。一边为了更好地提高自己的时间价值和生活质量而暗无天日的加班熬夜搞科研写论文，一边耗费大量时间做自己不擅长的事情上面，然后安慰自己今天终于把论文搞定，还节省了不少翻译费和时间。找鑫宇，帮翻译。省心客服QQ：100576934 ③鑫宇荐稿 EI&ISTP国际会议荐稿，祝您快速发表检索论文往年差几篇检索和核心论文影响职称评审和工资，延迟毕业，一年下来损失好几万，今年得赶紧了，要不然再等一年，代价太大了。可是投稿会议马上快要截稿了，不知道是否来得及，审稿是否能够通过，想赶紧搞定，怎么办？找鑫宇，帮荐稿。省心客服QQ：100576934 ④鑫宇排版论文排版，您的时间更有专业价值中文论文写好了，还要按照模板文档画图排版，自己对排版不专业，有些软件还不会用，画图太耗时间了，找打印店人家都嫌麻烦，怎么办？找鑫宇，帮排版。倾心客服QQ：100476934 ⑤鑫宇检测论文学术检测服务，确保论文学术质量中文论文写好了，引用了不少参考文献，想发表又担心其中有些引用的学术问题，将来万一被抓住把柄，怎么办？找鑫宇，帮检测。倾心客服QQ：100476934 ⑥鑫宇下载参考文献下载服务，您的论文依据更加充分经常写论文，经常需要阅读各种参考文献，家里又没有教育网不能免费下载文献，怎么办？找鑫宇，帮下载。倾心客服QQ：100476934 ⑦鑫宇查检论文检索查询，检索结果早知道去年发表的论文，说是3-6个月检索，到现在还没有消息，到底检索了没有？找鑫宇，帮查检。倾心客服QQ：100476934 ⑧鑫宇检索论文检索报告，教育部查新站红章权威证明每年评职称，写基金都要交各种学术科研成果证明，总是抱一大摞几十斤的书本期刊论文集去打字复印店。有没有其实什么证明学术科研成果。鑫宇检索淘宝首家，销量排名第一。教育部查新站红章权威证明。找鑫宇，帮证明。省心客服QQ：100576934 翻译找鑫宇，EI很简单！！您交给鑫宇中文，鑫宇为您终极搞定论文检索。 1）时间进度，让您省心，一般情况下3-5个工作日交付您画图排版好的译稿。 2）翻译质量，让您放心，50个专八译员随时待命，超过200万字的翻译经验。 3）服务信誉，让您舒心，淘宝交易，不满意不付款。 4）服务价格，让您开心。鑫宇服务费用除普通注册费外，仅1000元而已。没错，就是1000元，您可以享有的超值服务，包含至少750元的专八翻译（150-180元千字符）+200元荐稿+20元排版 +10元提交 +10元注册 +50元检索报告）！！！您再看看网络上随便找个论文中介机构，咨询一下一般期刊，核心期刊的代理发表价格，仅仅是推荐中文期刊发表，代理费（版面费）就是1000多，而鑫宇包含翻译-排版-提交-荐稿-注册-检索在内的总费用，仅是其他论文中介机构一般中文期刊的标准，甚至更便宜，而我们提供全面、系统、贴心的一条龙服务。您也可以在看看网络上科技论文润色和翻译的组织，通常0.20-0.40/字。除对您的学术生涯负责不提供中文代写外，鑫宇为您提供完整的论文三大检索解决方案，高效快速提升的学术影响力和学术成果。您交给鑫宇中文，鑫宇交给您终极检索证明。正如鑫宇众星拱月LOGO涵义：鑫宇，借“心语”用心与客户对话，为客户提供省时省事更省心的贴心服务。鑫宇，亦即“信誉”的谐音，诚信服务，信誉是生存之道，立业之本。不同颜色箭头，指向核心圆，意即众星拱月，象征鑫宇提供平台，通过各层次、各方面、整合资源多种方式通力合作，为客户提供专业以外的增值服务，让客户省时省事省心，更多关注专业核心！现在，您就可以在淘宝上找找看： 1）有没有从中文-翻译-排版-提交-荐稿-注册-三大检索报告全套服务的商家？ 2）有没有专职客服，一对一跟踪论文服务，随时通知进度的商家？ 3）有没有敢承诺投诉减免甚至全额退款的EI检索、论文翻译服务的商家？ 4）有没有全套服务(专八翻译+排版+提交+荐稿+注册+专职客服+投诉减免）收1000元服务费的商家？至少我们目前还没有找到，如果您找到了，请随时告诉我们！鑫宇特色 1）淘宝首推EI检索报告 EI检索证明排名第一 2）淘宝首推EI国际会议超市模式，按专业按主题列表供您自选鑫宇信誉鑫宇2010年-10-28日淘宝上线，600多用户，30%回头客。尽管知道网络上有很多刷钻的捷径，我们仍然坚守拓荒主义，都是一笔一笔的攒下来的，下面就是近期交易截图。细心的亲友们会发现，鑫宇的冲钻有很多是老客户的支持。没错，您只需要在普通注册费外加付1000元服务费，而您将获得的超值套餐服务，包括： 1) 750元以上的专八翻译费（150-180每千字符） 2) 20元排版费 3) 10元提交费 4）200元荐稿费 4) 10元注册费 5) 50元检索报告费（含快递）请您再次清点确认。您可能还在犹豫，到底值不值啊？翻译质量到底行不行啊？你们翻译的论文能不能及时录用啊？？录用了能不能检索啊？？？如果您有的是时间，我们相信您完全有能力可以自己全部搞定文献下载，论文检测，翻译，排版，提交，注册，检索，或者分单项找其他服务商。欢迎您咨询客服寻求满意答案或访问我们的网站。http://xinyustudio.taobao.com 至于您信不信，反正我是信了

1 0