Re: Article on multi-code processors - does it apply to z196?
That article is merely a popularist comment on the original paper. 'nuff said. A quick read of the original paper shows it to be both current (2.6.35-rc3 kernel - for which they also developed some fixes) and thorough. To some extent it was artificial in that they constructed the tests to do no physical I/O, but is interesting none-the-less. Given that Linux is (NUMA) node aware, it would be interesting to see how a (non-z/VM) multi-book s390x Linux partition would appear re node numbers and scheduler queues. My expectation would be that each book would appear as a node - with its attendant memory. I have no idea how z/VM may mangle this representation for a similarly defined guest. There used to be a fella named Alan around these parts that might have known ... Shane ... On Thu, 30 Sep 2010 12:01:33 -0500 McKown, John wrote: http://www.conceivablytech.com/3166/science-research/current-operating-systems-may-only-make-sense-up-to-48-cores/ Of course, the article is all about Intel and Linux. But I wonder if similar problems could occur on the 80 CP z196. If not, that would be a boon argument to replace largely multi-core Intel with z! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Article on multi-code processors - does it apply to z196?
Colleague who are interested in this subject may want to check the following latest R D Journal Issue by IBM Research And Development Regards Meral http://ieeexplore.ieee.org/xpl/tocresult.jsp?isnumber=5571943 Commercial Software for Multicore Systems Commercial Software for Multicore Systems Microprocessor and computing-system designs are making increasing use of parallelism, which includes homogeneous and heterogeneous chip multiprocessors, or multicores, as a means of providing continuing improvement in performance and efficiency. This issue presents key aspects of software for parallel computing and multicore systems in the commercial domain. This issue concludes with two nontopical papers. Topics include a subsystem for enhancing operating system health in the cloud computing era and the Open Computing Language (OpenCL) for use in digital TV applications. Preface B. Blainey, H. Franke, and M. Hind, Guest Editors 1 Exploitation of multicore systems in a Java virtual machine R. A. Sciampacone, V. Sundaresan, D. Maier, and T. Gray-Donald 2 Observations on tuning a Java enterprise application for performance and scalability E. Altman, M. Arnold, R. Bordawekar, R. M. Delmonico, N. Mitchell, and P. F. Sweeney 3 Detection of deadlock potentials in multithreaded programs R. Agarwal, S. Bensalem, E. Farchi, K. Havelund, Y. Nir-Buchbinder, S. D. Stoller, S. Ur, and L. Wang 4 THOR: A performance analysis tool for Java applications running on multicore systems Q. M. Teng, H. C. Wang, Z. Xiao, P. F. Sweeney, and E. Duesterwald 5 A taxonomy of accelerator architectures and their programming models C. Cazcaval, S. Chatterjee, H. Franke, K. J. Gildea, and P. Pattnaik -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Shane Sent: Friday, October 01, 2010 10:57 AM To: IBM-MAIN@bama.ua.edu Subject: Re: [IBM-MAIN] Article on multi-code processors - does it apply to z196? That article is merely a popularist comment on the original paper. 'nuff said. A quick read of the original paper shows it to be both current (2.6.35-rc3 kernel - for which they also developed some fixes) and thorough. To some extent it was artificial in that they constructed the tests to do no physical I/O, but is interesting none-the-less. Given that Linux is (NUMA) node aware, it would be interesting to see how a (non-z/VM) multi-book s390x Linux partition would appear re node numbers and scheduler queues. My expectation would be that each book would appear as a node - with its attendant memory. I have no idea how z/VM may mangle this representation for a similarly defined guest. There used to be a fella named Alan around these parts that might have known ... Shane ... On Thu, 30 Sep 2010 12:01:33 -0500 McKown, John wrote: http://www.conceivablytech.com/3166/science-research/current-operating-systems-may-only-make-sense-up-to-48-cores/ Of course, the article is all about Intel and Linux. But I wonder if similar problems could occur on the 80 CP z196. If not, that would be a boon argument to replace largely multi-core Intel with z! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html This message and attachments are confidential and intended solely for the individual(s) stated in this message. If you received this message although you are not the addressee, you are responsible to keep the message confidential. The sender has no responsibility for the accuracy or correctness of the information in the message and its attachments. Our company shall have no liability for any changes or late receiving, loss of integrity and confidentiality, viruses and any damages caused in anyway to your computer system. Bu mesaj ve ekleri, mesajda gonderildigi belirtilen kisi/kisilere ozeldir ve gizlidir. Bu mesajin muhatabi olmamaniza ragmen tarafiniza ulasmis olmasi halinde mesaj iceriginin gizliligi ve bu gizlilik yukumlulugune uyulmasi zorunlulugu tarafiniz icin de soz konusudur. Mesaj ve eklerinde yer alan bilgilerin dogrulugu ve guncelligi konusunda gonderenin ya da sirketimizin herhangi bir sorumlulugu bulunmamaktadir. Sirketimiz mesajin ve bilgilerinin size degisiklige ugrayarak veya gec ulasmasindan, butunlugunun ve gizliliginin korunamamasindan, virus icermesinden ve bilgisayar sisteminize verebilecegi herhangi bir zarardan sorumlu tutulamaz. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Article on multi-code processors - does it apply to z196?
I think those are not a single CEC, but multiple independent CECs perhaps more like a Parallel Sysplex in z/OS terms. From: IBM Mainframe Discussion List [ibm-m...@bama.ua.edu] On Behalf Of Tom Marchant [m42tom-ibmm...@yahoo.com] Sent: Thursday, September 30, 2010 12:29 PM To: IBM-MAIN@bama.ua.edu Subject: Re: Article on multi-code processors - does it apply to z196? On Thu, 30 Sep 2010 12:01:33 -0500, McKown, John wrote: http://www.conceivablytech.com/3166/science-research/current-operating-systems-may-only-make-sense-up-to-48-cores/ Of course, the article is all about Intel and Linux. What about the massively parallel systems running Linux using hundreds or thousands of Intel processors? -- Tom Marchant -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Article on multi-code processors - does it apply to z196?
http://www.conceivablytech.com/3166/science-research/current-operating-systems-may-only-make-sense-up-to-48-cores/ Of course, the article is all about Intel and Linux. But I wonder if similar problems could occur on the 80 CP z196. If not, that would be a boon argument to replace largely multi-core Intel with z! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Article on multi-code processors - does it apply to z196?
It reminds me of an xkcd http://xkcd.com/619/ I don't think it will be a real selling point, when you are comparing similar class machines. Kernel modifications have been done to address this on specific Linux systems. However if you were to use it as a pointer to why this class of machine is better than the one step lower, that is another story. -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of McKown, John Sent: Thursday, September 30, 2010 12:02 PM To: IBM-MAIN@bama.ua.edu Subject: Article on multi-code processors - does it apply to z196? http://www.conceivablytech.com/3166/science-research/current-operating-systems-may-only-make-sense-up-to-48-cores/ Of course, the article is all about Intel and Linux. But I wonder if similar problems could occur on the 80 CP z196. If not, that would be a boon argument to replace largely multi-core Intel with z! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Article on multi-code processors - does it apply to z196?
On Thu, 30 Sep 2010 12:01:33 -0500, McKown, John wrote: http://www.conceivablytech.com/3166/science-research/current-operating-systems-may-only-make-sense-up-to-48-cores/ Of course, the article is all about Intel and Linux. What about the massively parallel systems running Linux using hundreds or thousands of Intel processors? -- Tom Marchant -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Article on multi-code processors - does it apply to z196?
John, Take a look at Microsoft's Midori Operating System initiative. Also, Amdahl's Law in the multicore era. Harry Date: Thu, 30 Sep 2010 12:01:33 -0500 From: john.mck...@healthmarkets.com Subject: Article on multi-code processors - does it apply to z196? To: IBM-MAIN@bama.ua.edu http://www.conceivablytech.com/3166/science-research/current-operating-systems-may-only-make-sense-up-to-48-cores/ Of course, the article is all about Intel and Linux. But I wonder if similar problems could occur on the 80 CP z196. If not, that would be a boon argument to replace largely multi-core Intel with z! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Article on multi-code processors - does it apply to z196?
Tom, Most of those are actually clusters of nodes with only a few cores, with each node having it's own memory and OS instance. So, it's not one computer with thousands of CPUs, but a network of thousands of computers, each having 2-8 CPUs. On Thu, Sep 30, 2010 at 1:29 PM, Tom Marchant m42tom-ibmm...@yahoo.comwrote: On Thu, 30 Sep 2010 12:01:33 -0500, McKown, John wrote: http://www.conceivablytech.com/3166/science-research/current-operating-systems-may-only-make-sense-up-to-48-cores/ Of course, the article is all about Intel and Linux. What about the massively parallel systems running Linux using hundreds or thousands of Intel processors? -- Tom Marchant -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html CONFIDENTIALITY/EMAIL NOTICE: The material in this transmission contains confidential and privileged information intended only for the addressee. If you are not the intended recipient, please be advised that you have received this material in error and that any forwarding, copying, printing, distribution, use or disclosure of the material is strictly prohibited. If you have received this material in error, please (i) do not read it, (ii) reply to the sender that you received the message in error, and (iii) erase or destroy the material. Emails are not secure and can be intercepted, amended, lost or destroyed, or contain viruses. You are deemed to have accepted these risks if you communicate with us by email. Thank you. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html