From gmercier@mail.med.upenn.edu Thu Sep 25 12:26:44 1997 Received: from mail.med.upenn.edu for gmercier@mail.med.upenn.edu by www.ccl.net (8.8.3/950822.1) id LAA13829; Thu, 25 Sep 1997 11:51:55 -0400 (EDT) Received: (from gmercier@localhost) by mail.med.upenn.edu (8.8.6/8.8.6) id LAA03240 for chemistry@www.ccl.net; Thu, 25 Sep 1997 11:51:50 -0400 (EDT) From: "Gustavo A. Mercier Jr" Message-Id: <199709251551.LAA03240@mail.med.upenn.edu> Subject: Supercomputer Grants? To: chemistry@www.ccl.net Date: Thu, 25 Sep 1997 11:51:49 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23-upenn3.1] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Hi! I have a project that is a candidate for supercomputer level computer power. Although I did use the Cornell Center while a graduate student (in the early days when centers where not saturated!), I would like info on the status of the currently active centers. The goal is to obtain long term trajectories (ca. 5ns) using rigid body dynamic simulations as implemented in Moldy. The molecules of interest are the metalloporphyrins. This is part of an ongoing project on the structure - relaxivity relationship of these molecules that are potential MRI contrast agents. From kim@traj.chem.wayne.edu Wed Sep 24 18:26:25 1997 Received: from traj.chem.wayne.edu for kim@traj.chem.wayne.edu by www.ccl.net (8.8.3/950822.1) id RAA09999; Wed, 24 Sep 1997 17:31:16 -0400 (EDT) Received: by traj.chem.wayne.edu (AIX 3.2/UCB 5.64/4.03) id AA16736; Wed, 24 Sep 1997 17:31:16 -0400 Date: Wed, 24 Sep 1997 17:31:14 -0400 (EDT) From: Kim Bolton To: chemistry@www.ccl.net Cc: gmercier@mail.med.upenn.edu Subject: PENTIUM PCs Message-Id: Mime-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="-1915151607-2078917053-875136674=:15959" This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. Send mail to mime@docserver.cac.washington.edu for more info. ---1915151607-2078917053-875136674=:15959 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear CCLers, I have received many responses to the following question that I posted on the CCL. My summary is given below, and the actual responses (including those for a similar question from Peter Schreiner) are attached. >We are planning to buy a Pentium PC that will be used for number >crunching only (i.e., it will *not* be used as a terminal). We plan >to incorporate the LINUX operating system. Molecular dynamics code >(VENUS96), semiemprical codes (e.g., MOPAC) and ab initio code >(Gaussian) will be run on the machine. Does anyone have experience >with these machines, so that they can comment on > >i) The model of Pentium PC that we need, or would prefer, to run the >above packages, especially Gaussian (e.g., Pentium II, Pentium Pro). > >ii) Is there is a distinct advantage in having the dual CPU machine, >or is it better to stick with a single precessor PC. > >iii) The amount of RAM and hard disk that is required. > >iv) Problems with fortran compilation on these PCs, and what fortran >compiler is preferred. > >v) Any other machines that may be better suited for our applications. > >I will summarise the responses, > >Many thanks >Kim > Summary: There is not much difference in performances of the Pentium Pro and Pentium II, and all such differences seem to depend on the type of calculation (the Pentium II is basically a Pentium Pro with MMX, and thus offers no improvement under the Linus operating system or when doing floating point intensive calculations). The recommendation is to buy the cheapest Pentium PC with a reasonable cache, RAM and hard disk. To run MD and electronic structure theory programs efficiently, one should have at least 64MB RAM, as large an L2 cache as possible, and about 5 GB of SCSI hard disk. The dual processor PCs appear to offer better performance per dollar although they are slightly slower than running two separate single processor PCs. Most respondents use the f2c/gcc compiler, although there are a few problems associated with it. Many thanks to all of those who responded to my question, Kim. (kim@traj.chem.wayne.edu) ---1915151607-2078917053-875136674=:15959 Content-Type: TEXT/PLAIN; charset=US-ASCII; name=CCLPENTIUM Content-Transfer-Encoding: BASE64 Content-ID: Content-Description: LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0NCg0KTWVzc2FnZXMgZnJvbSBDQ0wgIA0KDQpJ IHRlc3RlZCBjYWxjdWxhdGlvbnMgdXNpbmcgR2F1c3NpYW4gOTQsIFJldmlz aW9uIEQuNA0KSW4gSW50ZWwgbWFjaGluZSwgSSB1c2VkIExpbnV4IEtlcm5l bCAyLjAuMzAgYXMgT1MuDQoNCjEuICNOIGIzcDg2L2dlbiA1ZCBmcmVxIDog Q2xPMiBjYWxjdWxhdGlvbiB1c2luZyA3OSBiYXNpcw0KICBQZW50aXVtIFBy byAyMDBNSHogOiAyNTYgbWlucw0KICBQZW50aXVtIElJICAyMzNNSHogOiAx OTIgbWlucw0KICBDcmF5VDNFIChERUMgQWxwaGEgRVY1LjYgNDUwTUh6LCA5 MDBNRkxPUFMsIG5vdCBwYXJhbGxlbGl6ZWQpIDogMzkgbWlucw0KDQoyLiAj TiBiM3A4Ni9nZW4gNWQgb3B0PXotbWF0cml4IDogQ2wyTyBjYWxjdWxhdGlv biB1c2luZyAxMzEgYmFzaXMNCiAgUGVudGl1bSAgICAgMTMzTUh6IDogMzM2 IG1pbnMNCiAgUGVudGl1bSBQcm8gMjAwTUh6IDogMTQ5IG1pbnMNCiAgUGVu dGl1bSBJSSAgMjMzTUh6IDogMTMwIG1pbnMNCiAgQ3JheVQzRSAoREVDIEFs cGhhIEVWNS42IDQ1ME1IeiwgOTAwTUZMT1BTLCBub3QgcGFyYWxsZWxpemVk KSA6IDU2IG1pbnMNCg0KLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0NClNv biwgU2FuZy1raWwgICAgICAgICAgICAgICAgICAgemFubmF2aUBvcmJpdC5r YWlzdC5hYy5rciAgICAgXFxfX19fLy8NCiAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICgu IC4pIHwNClF1YW50dW0gQ2hlbWlzdHJ5IExhYm9yYXRvcnkgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgIHwgIiB8IHwNCkRlcHQuIG9mIENo ZW1pc3RyeSwgS0FJU1QsIFNvdXRoIEtvcmVhICAgICAgICAgICAgICAgICAg ICAgICBDQ3xfX18vQ0NDDQogICAgICAgICAgICAgICAgICAgICAgICBodHRw Oi8vY21zcmlzYy5rYWlzdC5hYy5rci9xY2xhYi8gICAgICAgIFUNCi0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tDQoNCkkgaGF2ZSBhIGJpdCBvZiBleHBl cmllbmNlIHdpdGggdGhpcy4gIEkgd291bGQgZGVmaW5pdGVseSBnbyBQZW50 aXVtDQpQcm8gaW5zdGVhZCBvZiBQZW50aXVtIElJLCBiYXNlZCB1cG9uIGJl bmNobWFya3MgSSd2ZSBzZWVuLg0KDQpEdWFsIHByb2Nlc3NvciBtb3RoZXJi b2FyZCBpcyBuaWNlLCBhbmQgY2FuIHNhdmUgbW9uZXkgb24gYSBwZXItY3B1 DQpiYXNpcy4gIE9uIHRoZSBvdGhlciBoYW5kLCBpZiB0aGUgcHJvY2Vzc29y cyBzaGFyZSBjYWNoZSBtZW1vcnksIGFuZA0KeW91ciBzb2Z0d2FyZSBtYWtl cyBoZWF2eSB1c2Ugb2YgY2FjaGUsIEkgdGhpbmsgdGhhdCB0aGUgcHJvY2Vz c29ycw0KY291bGQgc2xvdyBlYWNoIG90aGVyIGRvd24uDQoNCldlIGVuZGVk IHVwIGNvbXBpbGluZyBmb3J0cmFuIGNvZGUgYnkgZmlyc3QgcnVubmluZyBp dCB0aHJvdWdoIGYyYw0KKGNvbnZlcnNpb24gdG8gYyBjb2RlISkgYW5kIHRo ZW4gcnVubmluZyB0aGUgYyBjb21waWxlci4gIFRoYXQNCmFjdHVhbGx5IHdv cmtlZCB3ZWxsLCB0aG91Z2ggdGhlcmUgd2VyZSBzb21lIGFubm95aW5nIGds aXRjaGVzIHRoYXQNCm5lZWRlZCB0byBiZSBkZWFsdCB3aXRoLg0KDQpJIHBy ZXN1bWUgeW91J3ZlIGNoZWNrZWQgdGhlIGJlbmNobWFya3MgdGhhdCBwZW9w bGUgaGF2ZSBwdWJsaXNoZWQgb24NCnRoZSB3ZWI/ICBJJ3ZlIGZvdW5kIHRo YXQgdGhlIHBlcmZvcm1hbmNlIG9mIG91ciBsaW51eCBkdWFsIHBlbnRpdW0N CnBybyBtYWNoaW5lIGRlcGVuZHMgc3Ryb25nbHkgdXBvbiB0aGUgdHlwZSBv ZiBjb2RlLiAgT25lIGFwcGxpY2F0aW9uDQpyYW4gZXh0cmVtZWx5IGZhc3Q7 IGFub3RoZXIgcXVpdGUgc2xvd2x5LiBJIGRvbid0IGZ1bGx5IHVuZGVyc3Rh bmQgd2h5DQp0aGV5IHNob3dlZCBzdWNoIGRpZmZlcmVudCBwZXJmb3JtYW5j ZS4gIEJ1dCBpZiB5b3UgYXJlIGdvaW5nIHRvIHJlbHkNCm9uIHRoaXMgbWFj aGluZSwgaXQgd291bGQgYmUgYmVzdCBpZiB5b3UgY291bGQgcnVuIHNvbWUg dGVzdHMgaW4NCmFkdmFuY2UuDQoNCmdvb2QgbHVjaw0KDQptaWtlDQotLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0NCk1pY2hhZWwgSy4gR2ls c29uLCBQaC5ELiwgTS5ELiAgICAgICAgICAgICAgICAgICAgICAgICBQaG9u ZTogICAoMzAxKSA3MzgtNjIxNw0KQ2VudGVyIGZvciBBZHZhbmNlZCBSZXNl YXJjaCBpbiBCaW90ZWNobm9sb2d5ICAgICAgICAgIEZheDogICAgICgzMDEp IDczOC02MjU1DQpOYXRpb25hbCBJbnN0aXR1dGUgb2YgU3RhbmRhcmRzIGFu ZCBUZWNobm9sb2d5DQo5NjAwIEd1ZGVsc2t5IERyaXZlICAgICAgICAgICAg IGh0dHA6Ly9pbmRpZ28xNS5jYXJiLm5pc3QuZ292L2NhcmIvZ2lsc29uLmh0 bWwNClJvY2t2aWxsZSwgTUQgIDIwODUwLTM0NzkgICAgICAgZS1tYWlsOiBn aWxzb25AaW5kaWdvMTQuY2FyYi5uaXN0Lmdvdg0KLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tDQoNCkRlYXIgQ0NMZXJzLA0KDQpUaGUgcXVl c3Rpb24NCg0KIkRvZXMgYW55b25lIGhhdmUgZXhwZXJpZW5jZSB3aXRoIG51 bWVyaWNhbCBhcHBsaWNhdGlvbnMgKEFiIGluaXRpbw0KY29kZXMgZXRjLikg b24gUGVudGl1bSBQcm8gdnMuIFBlbnRpdW0gSUkgdW5kZXIgTGludXggc3lz dGVtcz8iDQp0cmlnZ2VyZWQgcXVpdGUgYSBudW1iZXIgb2YgdXNlZnVsIHJl c3BvbnNlcy4gSXQgbG9va3MgbGlrZSBQQ3MgYXJlDQpmaW5hbGx5IHJlcGxh Y2luZyBhdmVyYWdlIHR5cGUgd29ya3N0YXRpb25zLiBJdCBpcyBzdGlsbCBu b3QgY2xlYXINCndoZXRoZXIgb3Igbm90IHRoZSBQZW50aXVtIElJIHNlcmll cyByZWFsbHkgaXMgYmV0dGVyIHRoYXQgdGhlIFBlbnRpdW0NClByb3MgLSBh dCBsZWFzdCB0aGV5IGFyZSBkaWZmZXJlbnQgaW4gcHJpY2UuDQoNCkhlcmUg YXJlIHRoZSByZXNwb25zZXM6DQoNCi0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tDQoNCllvdSBj YW4gZmluZCBzb21lIHRpbWluZ3MgY29uY2VybmluZyBQQyBHQU1FU1MgYmVu Y2htYXJrcyBvbiBQZW50aXVtIFBybyBhbmQNClBlbnRpdW0gSUkgc3lzdGVt cyB1bmRlciBXaW5kb3dzIE5UIG9uIHRoZSBVUkxzOg0KDQpodHRwOi8vY2xh c3NpYy5jaGVtLm1zdS5zdS9ncmFuL2dhbWVzcy9pbmRleC5odG1sDQpodHRw Oi8vY2xhc3NpYy5jaGVtLm1zdS5zdS9ncmFuL2dhbWVzcy9wZXJmb3JtYW5j ZS5odG1sDQoNCk1heSBiZSB0aGVzZSBjYW4gc29tZXdoYXQgaGVscCB0byBl c3RpbWF0ZSBwZXJmb3JtYW5jZSB1bmRlciBMaW51eC4NCg0KUmVnYXJkcywN Cg0KICAgIEFsZXguIEEuIEdyYW5vdnNreQ0KICAgIE1vc2NvdyBTdGF0ZSBV bml2ZXJzaXR5DQoNCi0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tDQpncmFuQGNsYXNzaWMuY2hl bS5tc3Uuc3UNCg0KJSUlJSUlJSUlDQoNCkkgYW0gcmVhbGx5IGludGVyZXN0 ZWQgaW4gYW55IGluZm9ybWF0aW9ucyB5b3Ugd2lsbCByZWNlaXZlLg0KTXkg cGVyc29uYWwgZXhwZXJpZW5jZSwgdXAgdG8gbm93LCBpcyBvbmx5IHdpdGgg dGhlIFBybyAyNTZrYiBjYWNoZS4NCkkgdXNlZCBpdCB0byBydW4gREZUIHBy b2dyYW1zLCBhbmQgdGhlIHBlcmZvcm1hbmNlLCB3aXRoIGJpZyBzeXN0ZW1z DQooMzAgYXRvbXMgb3IgbW9yZSkgYXJlIGV4Y2VsbGVudC4gSXQgaXMgODUl IG9mIGFuIElCTSByczYwMDAgTW9kLiAzQ1QNCndpdGggMU1CIG9mIEwyIGNh Y2hlIG1lbW9yeS4gQW5kIGl0IGNvc3RzIHJvdWdobHkgMjAlLiBSZW1hcmth YmxlIQ0KDQpBcyBmb3IgdGhlIFBJSSwgSSBoYXZlIG5vIGV4cGVyaWVuY2Uu IEFueWhvdywgSSAidGhpbmsiIHRoYXQgdGhlIFBybw0KZXZlbiBpZiB0aGUg Y2xvY2sgaXMgc2xvd2VyLCBzaG91bGQgcGVyZm9ybSBiZXR0ZXIuIFRoZSBM MiBjYWNoZSBpbg0KdGhlIFBybyBpcyBmdWxseSBpbnRlZ3JhdGVkIG9uIHRo ZSBwcm9jZXNzb3IgYW5kIHRodXMgcnVuIGF0IHRoZQ0Kc2FtZSBzcGVlZCBv ZiB0aGUgcHJvY2Vzc29yIGl0c2VsZi4gIE9uIHRoZSBQSUksIGluc3RlYWQs IHRoZQ0KdGhlIEwyIG1lbW9yeSBydW5zIGF0IDEvMiBvZiB0aGUgc3BlZWQg b2YgdGhlIHByb2Nlc3Nvci4NCg0KRmluYWxseSwgdGhlIFBybyBpcyBzb21l d2hhdCBjaGVhcGVyIHRoYW4gdGhlIFBJSSAyNjYgTUh6Lg0KDQpIb3BlIGl0 IGhlbHBzLA0KQmVzdCByZWdhcmRzDQoNCkx1aWdpDQoNCkx1aWdpIENhdmFs bG8NCkRlcHQuIG9mIENoZW1pc3RyeQ0KVW5pdi4gb2YgTmFwbGVzLCBJVEFM WQ0KRW1haWwgY2F2YWxsb0BjaGVtbmEuZGljaGkudW5pbmEuaXQNCg0KDQol JSUlJSUlJSUlDQoNCldoaWxlIEkgZG9uJ3QgaGF2ZSBhbnkgdGltaW5ncyBv ciBiZW5jaG1hcmtzIGZvciBQZW50aXVtIElJIHN5c3RlbXMsDQpJIHN1Z2dl c3QgeW91IGhhdmUgYSBsb29rIGF0IGh0dHA6Ly93d3cuc3BlY2JlbmNoLm9y Zy8gZm9yIFNQRUNpbnQ5NSBhbmQNClNQRUNmcDk1IHJlc3VsdHMuICBGcm9t IHdoYXQgSSd2ZSBzZWVuIHRoZSBwZW50aXVtIElJIGlzIG11Y2ggYmV0dGVy IGluDQpmbG9hdGluZyBwb2ludCBhbmQgYmV0dGVyIGluIGludGVnZXIgdGhh biB0aGUgcGVudGl1bSBwcm8uDQoNCkFsc28gZGlyZWN0IHlvdXIgd2ViIGJy b3dzZXIgYXQNCmh0dHA6Ly93d3cuaW50ZWwuY29tL3Byb2NzL3BlcmYvUGVu dGl1bUlJL3NwZWM5NWludC5odG0NCmFuZA0KaHR0cDovL3d3dy5pbnRlbC5j b20vcHJvY3MvcGVyZi9QZW50aXVtSUkvY3B1bWFyazMyLmh0bQ0KYW5kDQpo dHRwOi8vd3d3LmludGVsLmNvbS9wcm9jcy9wZXJmL1BlbnRpdW1JSS9ub3J0 b25zaTMyLmh0bQ0KDQpUbyB0aGUgYm90dG9tIG9mIHRoaXMgZW1haWwgSSd2 ZSBhdHRhY2hlZCB0aGUgb3V0cHV0IG9mIGNwdSBiZW5jaG1hcmtzIEkgcmFu IG9uDQphIGR1YWwgUFBSTyAyMDAgKHRoZSBmYWN0IHRoYXQgaXQgaXMgZHVh bCBtZWFucyBub3RoaW5nOyB0aGUgU01QIGNvZGUgZm9yIExpbnV4DQppcyBi YWQgcmlnaHQgbm93IGFuZCB0aGUgYmVuY2htYXJrcyBhcmUgbm90IG11bHRp dGhyZWFkaW5nKS4NCg0KZ29vZCBsdWNrLA0KUmljaA0KDQpCWVRFbWFyayAo dG0pIE5hdGl2ZSBNb2RlIEJlbmNobWFyayB2ZXIuIDIgKDEwLzk1KQ0KTlVN RVJJQyBTT1JUICAgIDogIEl0ZXJhdGlvbnMvc2VjLjogICAgICAgICA3MC4y NCAgSW5kZXg6ICAgMS44MA0KU1RSSU5HIFNPUlQgICAgIDogIEl0ZXJhdGlv bnMvc2VjLjogICAgICAgICAgNC4zNSAgSW5kZXg6ICAgMS45NA0KQklURklF TEQgICAgICAgIDogIEl0ZXJhdGlvbnMvc2VjLjogICAxNjk3NjY5Mi41MiAg SW5kZXg6ICAgMi45MQ0KRlAgRU1VTEFUSU9OICAgIDogIEl0ZXJhdGlvbnMv c2VjLjogICAgICAgICAgMy45MCAgSW5kZXg6ICAgMS44Nw0KRk9VUklFUiAg ICAgICAgIDogIEl0ZXJhdGlvbnMvc2VjLjogICAgICAgMTM5My4yNiAgSW5k ZXg6ICAgMS41OA0KQVNTSUdOTUVOVCAgICAgIDogIEl0ZXJhdGlvbnMvc2Vj LjogICAgICAgICAgMC43MSAgSW5kZXg6ICAgMi43MQ0KSURFQSAgICAgICAg ICAgIDogIEl0ZXJhdGlvbnMvc2VjLjogICAgICAgIDEzOC4yOSAgSW5kZXg6 ICAgMi4xMg0KSFVGRk1BTiAgICAgICAgIDogIEl0ZXJhdGlvbnMvc2VjLjog ICAgICAgICA2My4xOSAgSW5kZXg6ICAgMS43NQ0KTkVVUkFMIE5FVCAgICAg IDogIEl0ZXJhdGlvbnMvc2VjLjogICAgICAgICAgMC43MSAgSW5kZXg6ICAg MS4xMw0KTFUgREVDT01QT1NJVElPTjogIEl0ZXJhdGlvbnMvc2VjLjogICAg ICAgICA1Mi44MSAgSW5kZXg6ICAgMi43NA0KLi4uZG9uZS4uLg0KPT09PT09 PT09PT1PVkVSQUxMPT09PT09PT09PT09PQ0KSU5URUdFUiBJTkRFWDogICAg ICAgICAgIDIuMTE5DQpGTE9BVElORy1QT0lOVCBJTkRFWDogICAgMS43MDAN CiAoOTAgTUh6IERlbGwgUGVudGl1bSA9IDEuMDAwKQ0KPT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PQ0KDQoNCiUlJSUlJSUlJSUlJSUlDQoNCiAg SSBkb24ndCBoYXZlIGFueSByZWFsIHF1YW50aXRhdGl2ZSBhbnN3ZXJzIGZv ciB5b3UsIGJ1dCB0aG91Z2h0IEkgbWlnaHQNCnBhc3MgYWxvbmcgbXkgZXhw ZXJpZW5jZSB3aXRoIHJ1bm5pbmcgR2F1c3NpYW4gOTQgdW5kZXIgTGludXgg b24gYSBzaW5nbGUNClBlbnRpdW0gUHJvIDIwME1IeiB3aXRoIDY0TUIgUkFN IGFuZCBhIGZhc3Qvd2lkZSBTQ1NJIGhhcmQgZHJpdmUuICBUaGUNCmNvdXBs ZSBvZiBydW5zIHRoYXQgSSBkaWQgdG8gY29tcGFyZSB0aGUgc3BlZWQgd2l0 aCBhbiBJbmRpZ28yIChzaW5nbGUNClI4MDAwIDc1TUh6IHByb2Nlc3Nvcikg c2VlbWVkIHRvIGdpdmUgYSBmYWN0b3Igb2YgYWJvdXQgMy00IGluIGNwdSB0 aW1lLg0KDQpUaGF0IGlzLCB0aGUgUGVudGl1bSBQcm8gMjAwIHdhcyAzLTQg dGltZXMgc2xvd2VyIHRoYW4gdGhlIFNHSS4gIENvbnNpZGVyaW5nDQp0aGUg cmVsYXRpdmUgY29zdHMgdGhpcyBpcyBwcmV0dHkgZ29vZCBpbiBteSBvcGlu aW9uLiAgSSBjb21waWxlZCBHYXVzc2lhbjk0DQp1c2luZyBmMmMvZ2NjLiAg SSBzb21ldGltZXMgd29uZGVyIGlmIEdhdXNzaWFuIHdvdWxkIHJ1biBmYXN0 ZXIgaWYgY29tcGlsZWQNCndpdGggYSBnb29kIGNvbW1lcmNpYWwgZm9ydHJh biBjb21waWxlciAoZS5nLiBBYnNvZnQpLiAgQnV0IGV2ZW4gdGhlDQplZHVj YXRpb25hbCBwcmljZSAoJDUyNC4yNSBVUykgaXMgdG9vIHByaWNleSBmb3Ig bWUgcmlnaHQgbm93Lg0KDQogIE15IGd1ZXNzIGlzIHRoYXQgeW91IHdpbGwg bm90IHNlZSBtdWNoIG9mIGFuIGltcHJvdmVtZW50IHdpdGggdGhlIFBlbnRp dW0gSUkNCmZvciB0aGVzZSB0eXBlcyBvZiBmbG9hdGluZyBwb2ludCBpbnRl bnNpdmUgY2FsY3VsYXRpb25zLiAgSSBiZWxpZXZlIHRoYXQNCm1vc3Qgb2Yg dGhlIGFkdmFudGFnZSBvZiB0aGUgUGVudGl1bSBJSSBjb21lcyBmcm9tIE1N WCBjYXBhYmlsaXRpZXMgd2hpY2gNCnlvdSB3aWxsIGZpbmQgdXNlbGVzcyB3 aXRoIExpbnV4LiAgQXQgbGVhc3QgdGhpcyBpcyB3aGF0IEkgcmVjYWxsIHJl YWRpbmcNCmluIExpbnV4IEpvdXJuYWwgb3IgVW5peCBSZXZpZXcuICBGbG9h dGluZyBwb2ludCBpcyBwcm9iYWJseSBvbmx5IG1hcmdpbmFsbHkNCmZhc3Rl ciB0aGFuIGEgUGVudGl1bSBQcm8uDQoNCiAgICAgIEhvcGUgdGhpcyBoZWxw cyAtIEpvaG4gQnVzaG5lbGwgICAgICBidXNobmVsbEBnYXVjaG8udWNzYi5l ZHUNCg0KDQolJSUlJSUlJSUlJSUlJQ0KDQpXZSBhcmUgaW1wbGVtZW50aW5n IGEgc3lzdGVtIGNhbGxlZCBMb0JvUywgc2hvcnQgZm9yIExvdHMgb2YgQm94 ZXMgb24NClNoZWx2ZXMsIHdoaWNoIGNvbnNpc3RzIG9mIDY0IGR1YWwgUGVu dGl1bSBQcm8gbWFjaGluZXMuICBXZSBvcmdhbml6ZSB0aGVtDQp0byBkbyBz dXBlcmNvbXB1dGluZyB3aXRoIG1vbGVjdWxhciBkeW5hbWljcyBjYWxjdWxh dGlvbnMuICBXZSB0ZXN0ZWQgdGhlDQpQZW50aXVtIFBybyBhbmQgdGhlIFBl bnRpdW0gSUkgcHJpb3IgdG8gc2VsZWN0aW5nIHRoZSBQZW50aXVtIFByby4N Cg0KDQoxLiBUaGUgUGVudGl1bSBQcm8gZ2l2ZXMgZXF1aXZhbGVudCBwZXJm b3JtYW5jZS4gIE91ciBiZW5jaG1hcmsgd2hpY2gNCiAgIGNvbnNpc3RzIG9m IGV4dGVuc2l2ZSBmbG9hdGluZyBwb2ludCAoQ0hBUk1NKSB0b29rIDIgaG91 cnMgb24gYSAyMDANCiAgIE1IeiBQZW50aXVtIFByby4gIE9uIGEgMjY2IE1I eiBQZW50aXVtIElJIHRoZSB0aW1lIHdhcyAyIGhvdXJzIHRpbWVzDQogICAo MjAwLzI2NikuDQoNClRoaXMgaXMgbm90IHN1cnByaXNpbmcgc2luY2UgYSBQ ZW50aXVtSUkgaXMgYSBQZW50aXVtIFBybyBjb3JlIHdpdGggdGhlDQpNTVgg aW5zdHJ1Y3Rpb25zIGFkZGVkIG9uLiAgTU1YLCB3aGljaCBpcyBlbnRpcmVs eSBpbnRlZ2VyIG9wZXJhdGlvbnMsDQpkb2Vzbid0IGRvIGFueXRoaW5nIGZv ciBmbG9hdGluZyBwb2ludCBvcGVyYXRpb25zLiAgSXQgbWlnaHQgZG8gc29t ZXRoaW5nDQpmb3IgcG9pbnRlciByZWZlcmVuY2VzIGFuZCBpbnRlZ2VyIG1h dGggYnV0IGNvbXBpbGVycyBkbyBub3QgY3VycmVudGx5IHVzZQ0KTU1YIGZv ciB0aGVzZSBvcGVyYXRpb25zLg0KDQpUaGUgUGVudGl1bUlJIGhhcyBzb21l IG5ldyBmZWF0dXJlcy4gIE9uZSB3aGljaCBpbXByb3ZlcyBwZXJmb3JtYW5j ZSBpcw0KYmV0dGVyIElPIHRvIHRoZSBzeXN0ZW0gYnVzLiAgSG93ZXZlciwg dGhpcyBpcyBlc3NlbnRpYWxseSBvZmZzZXQgYnkgdGhlDQpmYWN0IHRoYXQg dGhlIEwyIGNhY2hlIHJ1bnMgYXQgaGFsZiB0aGUgQ1BVIHNwZWVkLiAgRXNz ZW50aWFsbHkgd2Ugc2F3IGENCndhc2ggYmV0d2VlbiB0aGVzZSB0d28gZGVz aWduIGNoYW5nZXMuDQoNCkJ1eSB0aGUgUGVudGl1bSBQcm8gISAgSXQgZ2l2 ZXMgeW91IE1VQ0ggbW9yZSBmb3IgdGhlIG1vbmV5Lg0KDQpFcmljIEJpbGxp bmdzDQpMb0JvUyBQcm9qZWN0DQpOSUgNCg0KDQolJSUlJSUlJSUlJSUNCg0K DQo+RGVhciBjb2xsZWFndWVzLA0KPg0KPldlIGFyZSBwbGFubmluZyB0byBi dXkgYSBQZW50aXVtIFBDIHRoYXQgd2lsbCBiZSB1c2VkIGZvciBudW1iZXIN Cj5jcnVuY2hpbmcgb25seSAoaS5lLiwgaXQgd2lsbCAqbm90KiBiZSB1c2Vk IGFzIGEgdGVybWluYWwpLiAgV2UgcGxhbg0KPnRvIGluY29ycG9yYXRlIHRo ZSBMSU5VWCBvcGVyYXRpbmcgc3lzdGVtLiAgTW9sZWN1bGFyIGR5bmFtaWNz IGNvZGUNCj4oVkVOVVM5NiksIHNlbWllbXByaWNhbCBjb2RlcyAoZS5nLiwg TU9QQUMpIGFuZCBhYiBpbml0aW8gY29kZQ0KPihHYXVzc2lhbikgd2lsbCBi ZSBydW4gb24gdGhlIG1hY2hpbmUuICBEb2VzIGFueW9uZSBoYXZlIGV4cGVy aWVuY2UNCj53aXRoIHRoZXNlIG1hY2hpbmVzLCBzbyB0aGF0IHRoZXkgY2Fu IGNvbW1lbnQgb24NCg0KSSBndWVzcyBJIGhhdmUgc29tZSBzdHVmZiB0byBj b21tZW50IG9uIGFzIG15IGdyb3VwIHJlY2VudGx5IGxvb2tlZCBpbnRvDQp0 aGUgc2FtZSB0aGluZyB5b3UgYXJlIGFza2luZy4gIFJpZ2h0IG5vdyBJIGhh dmUgYSBtaWxsaW9uIHRoaW5ncyBnb2luZyBvbg0Kc28gSSBoYXZlbid0IGJl ZW4gYWJsZSB0byBzdW1tYXJpemUgbXkgZmluZGluZ3MgYW5kIHBvc3QgdGhl bSB0byB0aGUgQ0NMIG9uDQpzZXR0aW5nIHVwIGEgRHVhbCBQZW50aXVtIFBy byBzeXN0ZW0gdG8gY2Fycnkgb3V0IEdhdXNzaWFuIGNhbGNzLiAgVGhpcw0K c2VlbXMgdG8gYmUgcmVhbCBhY3RpdmUgYXJlYSB0aGVzZSBkYXlzLCBidXQg SSBhbSBub3Qgc3VyZSB0aGUgImFuc3dlciIgdG8NCmhvdyB3ZWxsIHRoZXNl IHN5c3RlbXMgd29yayByZWxhdGl2ZSB0byBvdGhlciBzZXR1cHMgaXMgaW4g eWV0LiAgTXlzZWxmIGFuZA0Kb3RoZXJzIChNYXJjIE5pY2tsYXVzIGF0IHRo ZSBOSUgpIGhhdmUgZm91bmQgc29tZSBpbnRlcmVzdGluZyBiZWhhdmlvciB3 aXRoDQp0aGVzZSBzeXN0ZW1zLCB3aGljaCBJIGFtIHN1cmUgd2UnbGwgYm90 aCByZXBvcnQgb24gd2hlbiB3ZSBmaW5pc2ggdXAgbW9yZQ0KdGVzdHMvaW52 ZXN0aWdhdGlvbnMuDQoNCkJvdHRvbSBsaW5lIHJpZ2h0IG5vdzogZ29pbmcg dGhpcyBQZW50aXVtL0xpbnV4IHJvdXRlIGlzIGEgcHJldHR5IGNoZWFwIHdh eQ0KdG8gIGhhdmUgYSBkZWNlbnQgd29ya3N0YXRpb24uDQoNCj4NCj5pKSBU aGUgbW9kZWwgb2YgUGVudGl1bSBQQyB0aGF0IHdlIG5lZWQsIG9yIHdvdWxk IHByZWZlciwgdG8gcnVuIHRoZQ0KPmFib3ZlIHBhY2thZ2VzLCBlc3BlY2lh bGx5IEdhdXNzaWFuIChlLmcuLCBQZW50aXVtIElJLCBQZW50aXVtIFBybyku DQoNCkkgdGhpbmsgYW55IFBDLCAzODYgYW5kIGFib3ZlLCBjYW4gcnVuIGxp bnV4LiAgSSBndWVzcyB0aGUgbmV3ZXIgUGVudGl1bQ0KSUkncyBhcmUgdGhl IGZhc3Rlc3Qgb25lcyBvdXQgdGhlcmUsIGFuZCB0aGUgZmFzdGVyIHRoZSBi ZXR0ZXIhICBBbg0KaW50ZXJlc3RpbmcgZmluZGluZyBJIGhhdmUgbm90aWNl ZCBpcyB0aGF0IHRoaXMgRHVhbCBQNiBJIGhhdmUgaXMgZmFzdGVyDQp0aGFu IG15IG1vZGVsIDU4MCByczYwMDAgb24gc21hbGwgbW9sZWN1bGVzLCBidXQg c2xvd3MgY29uc2lkZXJhYmx5DQpyZWxhdGl2ZSB0byBteSByczYwMDAgb24g bGFyZ2VyIG1vbGVjdWxlcy4gIEl0IHNlZW1zIHRoZXNlIFBlbnRpdW1zIGhh dmUNCiJtZW1vcnkgYmFuZHdpZHRoIiBwcm9ibGVtcyB0aGF0ICJyZWFsIiB3 b3Jrc3RhdGlvbnMgYXJlIG5vdCBzbyBwbGFndWVkDQp3aXRoLg0KDQo+DQo+ aWkpIElzIHRoZXJlIGlzIGEgZGlzdGluY3QgYWR2YW50YWdlIGluIGhhdmlu ZyB0aGUgZHVhbCBDUFUgbWFjaGluZSwNCj5vciBpcyBpdCBiZXR0ZXIgdG8g c3RpY2sgd2l0aCBhIHNpbmdsZSBwcmVjZXNzb3IgUEMuDQoNCkdvb2QgcXVl c3Rpb24sIHdoaWNoIEkgZG9uJ3QgaGF2ZSB0aGUgYW5zd2VyIHRvLiAgSSBk byBrbm93IHRoZXJlIGlzIGENCnBlcmZvcm1hbmNlIGhpdCBvZiBtYXliZSAx MC0yMCUgZm9yIHJ1bm5pbmcgdHdvIEc5NCBqb2JzIHNpbXVsdGFuZW91c2x5 IG9uDQpteSBEdWFsIFA2LCBidXQgSSBoYXZlIHRoaXMgZmVlbGluZyBpdCBt YXkgc3RpbGwgYmUgbW9yZSBlY29ub21pY2FsIHRvIGdvDQp0aGlzIHJvdXRl IHJhdGhlciB0aGFuIHR3byBzaW5nbGUgcHJvY2Vzc29yIG1hY2hpbmVzLCBi dXQgSSBoYXZlbid0DQppbnZlc3RpZ2F0ZWQgdGhpcyBhdCBhbGwuICBKdXN0 IGEgZ3Vlc3MuDQoNCj4NCj5paWkpIFRoZSBhbW91bnQgb2YgUkFNIGFuZCBo YXJkIGRpc2sgdGhhdCBpcyByZXF1aXJlZC4NCg0KTW9yZSB0aGUgYmV0dGVy LCBidXQgSSB3b3VsZCBndWVzcyB5b3UgY291bGQgZ2V0IGJ5IHdpdGggMzIg TWIgb2YgcmFtLiAgQW5kDQp0aGVzZSBkYXlzIHJhbSBpcyBwcmV0dHkgY2hl YXAgc28ganVzdCBhIGZldyBkb2xsYXJzIG1vcmUgeW91IGNhbiBiZSB1cCB0 bw0KMTI4LCBsaWtlIHdlIGRpZC4NCg0KPg0KPml2KSBQcm9ibGVtcyB3aXRo IGZvcnRyYW4gY29tcGlsYXRpb24gb24gdGhlc2UgUENzLCBhbmQgd2hhdCBm b3J0cmFuDQo+Y29tcGlsZXIgaXMgcHJlZmVycmVkLg0KDQpXZSBjb21waWxl ZCBHOTQgd2l0aCBmMmMgYW5kIGdjYy4gIEkgdGhpbmsgdGhhdCBpcyBhIHN0 YW5kYXJkIHdheSB0bw0KY29tcGlsZSBmb3J0cmFuIHByb2dyYW1zIG9uIGEg bGludXggbWFjaGluZSwgYW5kIGl0IG1vcmUgb2YgbGVzcyB3ZW50DQp3aXRo b3V0IGFueSBwcm9ibGVtcy4NCg0KPg0KPnYpIEFueSBvdGhlciBtYWNoaW5l cyB0aGF0IG1heSBiZSBiZXR0ZXIgc3VpdGVkIGZvciBvdXIgYXBwbGljYXRp b25zLg0KDQpJJ2xsIGJldCBvdGhlcnMgd2lsbCBnaXZlIHlvdSBpbmZvIG9u IHRoaXMgYmV0dGVyIHRoYW4gSSBjYW4uDQoNCkpvZQ0KDQoNCi0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0NCkpvc2VwaCBXLiBCYXVzY2gsIFBo LkQuDQpBc3Npc3RhbnQgUHJvZmVzc29yIG9mIENoZW1pc3RyeQ0KRGVwYXJ0 bWVudCBvZiBDaGVtaXN0cnkNClZpbGxhbm92YSBVbml2ZXJzaXR5DQpWaWxs YW5vdmEsIFBBICAxOTA4NS0xNjk5DQpWb2ljZTogIDYxMC01MTktNDg3Mg0K RkFYOiAgICA2MTAtNTE5LTYzNDcgKHByZWZlcnJlZCkgb3IgNjEwLTUxOS03 MTY3DQpFbWFpbDogIGJhdXNjaEByczZjaGVtLnZpbGwuZWR1DQpXV1c6ICAg IGh0dHA6Ly9yczZjaGVtLnZpbGwuZWR1L2ZhY3VsdHkvYmF1c2NoLmh0bQ0K LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLQ0KS2ltOg0KDQpJIGNh biBtYWtlIGEgZmV3IGNvbW1lbnRzIGFib3V0IHJ1bm5pbmcgR2F1c3NpYW4g dW5kZXIgTGludXggb24gcGVudGl1bXMuDQpEdXJpbmcgbXkgcG9zdGRvYywg d2Ugd2VyZSBydW5uaW5nIEc5NCBvbiB0d28gZGlmZmVyZW50IFBlbnRpdW0g UHJvIDIwMA0Kc3lzdGVtczogYSBzaW5nbGUgcHJvY2Vzc29yIHN5c3RlbSB3 aXRoIDY0IE1iIFJBTSBhbmQgYSA0R2IgaGFyZCBkaXNrIGFuZCBhDQpkdWFs IHByb2Nlc3NvciB3aXRoIDEyOCBNYiBSQU0gYW5kIGEgOSBHYiBkaXNrLg0K DQpUaGUgcGVyZm9ybWFuY2Ugb2YgYm90aCBvZiB0aGVzZSBtYWNoaW5lcyB1 bmRlciBMaW51eCAodmVyc2lvbiAyLjAuMCBJSVJDKQ0Kd2FzIGNvbXBhcmFi bGUgdG8gYSBzaW1sYXJseSBlcXVpcHBlZCB3b3Jrc3RhdGlvbiBhdCBhIGNv bnNpZGVyYWJseSByZWR1Y2VkDQpwcmljZS4gIFRoZSBzZWNvbmQgcHJvY2Vz c29yIGFjdHVhbGx5IG1ha2VzIGEgc2luZ2xlIGpvYiBydW4gc2xpZ2h0bHkN CmZhc3RlciB0aGFuIHRoZSBzaW5nbGUgcHJvY2Vzc29yIG1hY2hpbmUgZHVl IHRvIHNvbWUgbG9hZCBzaGFyaW5nLCBhbHRob3VnaA0KeW91IHByb2JhYmx5 IHdhbnQgdG8gcnVuIHR3byBqb2JzIHNpbXVsdGFuZW91c2x5IG9uIHRoZSBk dWFsIG1hY2hpbmUuDQoNCk9uZSBDYXZlYXQ6ICBhbmFseXRpY2FsIHZpYnJh dGlvbmFsIGZyZXF1ZW5jaWVzIHVzaW5nIGFueSBvZiB0aGUgREZUDQptZXRo b2RzIGluIEc5NCBhcmUgdW5yZWFzb25hYmx5IHNsb3cuICBJIG5ldmVyIHJl YWxseSByZWNlaXZlZCBhbiBhbnN3ZXINCmZyb20gR2F1c3NpYW4gb24gdGhp cyBwb2ludC4NCg0KQWxsIGluIGFsbCwgSSB3b3VsZCByZWNvbW1lbmQgdGhl c2Ugc3lzdGVtcy4gIFRoZXkgYXJlIGEgZ3JlYXQgY29tcHJvbWlzZQ0Kb2Yg cGVyZm9ybWFuY2UgYW5kIHByaWNlLg0KDQpHb29kIEx1Y2sNCkJpbGwNCg0K RHIuIFdpbGxpYW0gTS4gRGF2aXMNCk5TRVJDIFZpc2l0aW5nIEZlbGxvdw0K RFJFQSwgRG9ja3lhcmQgTGFib3JhdG9yeSBELTE3DQpQLk8uIEJveCAxMDEy DQpEYXJ0bW91dGgsIE5TIA0KQ0FOQURBIEIyWSAzWjcNClBob25lOiAgKDkw MikgNDI3LTA1NTAgRVhULiAzNDMxDQpGYXg6ICAgICg5MDIpIDQyNy0zNDM1 DQpkYXZpc3dAZHJlYS5kbmQuY2ENCg0KLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLQ0KDQoNCj5EZWFyIGNvbGxlYWd1ZXMsDQo+DQo+V2UgYXJl IHBsYW5uaW5nIHRvIGJ1eSBhIFBlbnRpdW0gUEMgdGhhdCB3aWxsIGJlIHVz ZWQgZm9yIG51bWJlcg0KPmNydW5jaGluZyBvbmx5IChpLmUuLCBpdCB3aWxs ICpub3QqIGJlIHVzZWQgYXMgYSB0ZXJtaW5hbCkuICBXZSBwbGFuDQo+dG8g aW5jb3Jwb3JhdGUgdGhlIExJTlVYIG9wZXJhdGluZyBzeXN0ZW0uICBNb2xl Y3VsYXIgZHluYW1pY3MgY29kZQ0KPihWRU5VUzk2KSwgc2VtaWVtcHJpY2Fs IGNvZGVzIChlLmcuLCBNT1BBQykgYW5kIGFiIGluaXRpbyBjb2RlDQo+KEdh dXNzaWFuKSB3aWxsIGJlIHJ1biBvbiB0aGUgbWFjaGluZS4gIERvZXMgYW55 b25lIGhhdmUgZXhwZXJpZW5jZQ0KPndpdGggdGhlc2UgbWFjaGluZXMsIHNv IHRoYXQgdGhleSBjYW4gY29tbWVudCBvbg0KPg0KPmkpIFRoZSBtb2RlbCBv ZiBQZW50aXVtIFBDIHRoYXQgd2UgbmVlZCwgb3Igd291bGQgcHJlZmVyLCB0 byBydW4gdGhlDQo+YWJvdmUgcGFja2FnZXMsIGVzcGVjaWFsbHkgR2F1c3Np YW4gKGUuZy4sIFBlbnRpdW0gSUksIFBlbnRpdW0gUHJvKS4NCj4NCg0KSSBh bSBjdXJyZW50bHkgcnVubmluZyBNT1BBQzcgYW5kIEdBTUVTUyBvbiBhIDEz M01IeiBQZW50aXVtLzEyOE1CDQpSQU0vNS41R0IgSEQvUmVkSGF0IDQuMSBM aW51eCBkaXN0cmlidXRpb24uICBNT1BBQzcgaW4gTGludXggcnVucyB+Mjg1 JQ0KZmFzdGVyIHRoYW4gTU9QQUM5NCBpbiBXaW45NSB3aXRoIHRoZSBzYW1l IHNldCBvZiBpbnB1dCBkYXRhLg0KDQpJbiBnZW5lcmFsLCBJIGFtIG5vdCBk aXNwbGVhc2VkIHdpdGggdGhlIHJlc3VsdHMgb2YgbXkgY29tcHV0YXRpb25h bCB3b3JrDQppbiBMaW51eC4gIEkgYW0gYWxzbyB1c2luZyBzZXZlcmFsIG1v bGVjdWxhciBtZWNoYW5pY3MgcHJvZ3JhbXMgcnVubmluZyBpbg0KKG9uID8p IExpbnV4Lg0KIA0KPmlpKSBJcyB0aGVyZSBpcyBhIGRpc3RpbmN0IGFkdmFu dGFnZSBpbiBoYXZpbmcgdGhlIGR1YWwgQ1BVIG1hY2hpbmUsDQo+b3IgaXMg aXQgYmV0dGVyIHRvIHN0aWNrIHdpdGggYSBzaW5nbGUgcHJlY2Vzc29yIFBD Lg0KPg0KDQo+aWlpKSBUaGUgYW1vdW50IG9mIFJBTSBhbmQgaGFyZCBkaXNr IHRoYXQgaXMgcmVxdWlyZWQuDQo+DQoNCkkgaHZlIGdvdHRlbiByZXN1bHRz IHdpdGggMzJNQiBvZiBSQU0gdGhhdCB3ZXJlIHNhdGlzZmFjdG9yeSwgIElu IGZhY3QgSQ0KanVzdCBpbmNyZWFzZWQgdGhlIFJBTSB0byAxMjlNQiBhIGNv dXBsZSBvZiB3ZWVrcyBhZ28uICBDYW4ndCByZWFsbHkNCmNvbW1lbnQgb24g aW5jcmVhc2UgcGVmb3JtYW5jZSB3aXRoIHJlZ2FyZCB0byBzcGVlZCBhdCB0 aGlzIHBvaW50LCBidXQNCmluY3JlYXNpbmcgUkFNIGNlcnRhaW5seSBpbmNy ZWFzZXMgdGhlIG1heGltdW0gc2l6ZSBvZiB0aGUgbW9sZWN1bGUuICBJDQpo YXZlIGhhZCBHQU1FU1Mgd3JpdGUgc2NyYXRjaCBmaWxlcyBhcyBsYXJnZSBh cyB+MTUwTUIgZHVyaW5nIGV4ZWN1dGlvbi4NCg0KPml2KSBQcm9ibGVtcyB3 aXRoIGZvcnRyYW4gY29tcGlsYXRpb24gb24gdGhlc2UgUENzLCBhbmQgd2hh dCBmb3J0cmFuDQo+Y29tcGlsZXIgaXMgcHJlZmVycmVkLg0KPg0KDQpJbml0 aWFsbHkgSSB3YXMgdXNpbmcgZ2NjLTIuNy4yL2YyYywgYnV0IGhhdmUgcmVj ZW50bHkgc3dpdGNoZWQgb3ZlciB0byB0aGUNCkdOVSBGNzcgY29tcGlsZXIu ICBmMmMgaXMgYSBzY3JpcHQgdGhhdCB0cmFuc2xhdGVzIHRoZSBGT1JUUkFO IHNvdXJjZSBjb2RlDQp0byBDIGFuZCB0aGVuIGRvZXMgdGhlIGNvbXBpbGlu ZyBhbmQgbGlua2luZy4gIEl0J3MgdGhlIG9ubHkgRk9SVFJBTg0KY29tcGls ZXIsIHNvIEknbSB0b2xkLCB0aGF0IElOU0lTVFMgb24gY29tcGxldGUgY29t cGxpYW5jZSB3aXRoIHRoZQ0KRk9SVFJBTjc3IHN0YW5kYXJkIC0gd2hpY2gg Y2FuIGxlYWQgdG8gYSBmZXcgbWlub3IgcHJvYmxlbXMuICBJIG11Y2ggcHJl ZmVyDQp0aGUgR05VIEY3NyBjb21waWxlciwgYnV0IGl0IGNhbiBiZSBhIGJp dCBxdWVya3kgdG8gZ2V0IHJ1bm5pbmcsIHRoYXQgZGlkDQpub3QsIGhvd2V2 ZXIgcHJvdmUgdG8gYmUgYSBwcm9ibGVtLg0KDQo+dikgQW55IG90aGVyIG1h Y2hpbmVzIHRoYXQgbWF5IGJlIGJldHRlciBzdWl0ZWQgZm9yIG91ciBhcHBs aWNhdGlvbnMuDQo+DQoNCkNhbid0IGFkZHJlc3MgdGhpcyBxdWVzdGlvbiwg SSB1c2Ugd2hhdCBJIGhhdmUNCg0KPkkgd2lsbCBzdW1tYXJpc2UgdGhlIHJl c3BvbnNlcywNCj4NCj5NYW55IHRoYW5rc0Zyb206ICJTdGVwaGVuIFAuIE1v bG5hciIgPHNtb2xuYXJAcG9zdGJveC5hY3Mub2hpby1zdGF0ZS5lZHU+DQoN Cj5LaW0NCg0KLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tDQoN Cj4gRGVhciBjb2xsZWFndWVzLA0KPiANCj4gV2UgYXJlIHBsYW5uaW5nIHRv IGJ1eSBhIFBlbnRpdW0gUEMgdGhhdCB3aWxsIGJlIHVzZWQgZm9yIG51bWJl cg0KPiBjcnVuY2hpbmcgb25seSAoaS5lLiwgaXQgd2lsbCAqbm90KiBiZSB1 c2VkIGFzIGEgdGVybWluYWwpLiAgV2UgcGxhbg0KPiB0byBpbmNvcnBvcmF0 ZSB0aGUgTElOVVggb3BlcmF0aW5nIHN5c3RlbS4gIE1vbGVjdWxhciBkeW5h bWljcyBjb2RlDQo+IChWRU5VUzk2KSwgc2VtaWVtcHJpY2FsIGNvZGVzIChl LmcuLCBNT1BBQykgYW5kIGFiIGluaXRpbyBjb2RlDQo+IChHYXVzc2lhbikg d2lsbCBiZSBydW4gb24gdGhlIG1hY2hpbmUuICBEb2VzIGFueW9uZSBoYXZl IGV4cGVyaWVuY2UNCj4gd2l0aCB0aGVzZSBtYWNoaW5lcywgc28gdGhhdCB0 aGV5IGNhbiBjb21tZW50IG9uDQo+IA0KPiBpKSBUaGUgbW9kZWwgb2YgUGVu dGl1bSBQQyB0aGF0IHdlIG5lZWQsIG9yIHdvdWxkIHByZWZlciwgdG8gcnVu IHRoZQ0KPiBhYm92ZSBwYWNrYWdlcywgZXNwZWNpYWxseSBHYXVzc2lhbiAo ZS5nLiwgUGVudGl1bSBJSSwgUGVudGl1bSBQcm8pLg0KDQo+IGlpKSBJcyB0 aGVyZSBpcyBhIGRpc3RpbmN0IGFkdmFudGFnZSBpbiBoYXZpbmcgdGhlIGR1 YWwgQ1BVIG1hY2hpbmUsDQo+IG9yIGlzIGl0IGJldHRlciB0byBzdGljayB3 aXRoIGEgc2luZ2xlIHByZWNlc3NvciBQQy4NCg0KTGludXggU01QIGlzIG5v dCB5ZXQgc3RhYmxlIGVub3VnaCB0byB1c2UgaXQgb24gYSBwcm9kdWN0aW9u IG1hY2hpbmUuDQpIb3dldmVyLCBpZiB5b3UgaGF2ZSBsaW51eCBmYW4gaW4g Z3JvdXAgYW5kIGNhbiBhbGxvdyByYXRoZXIgc21hbGwNCnVwdGltZXMgKGZy b20gZGF5cyB0byBmZXcgd2Vla3MpIC0tIHRoZW4gaXQgbWF5IGJlIGEgZ29v ZCBjaG9pY2UuDQogDQo+IGlpaSkgVGhlIGFtb3VudCBvZiBSQU0gYW5kIGhh cmQgZGlzayB0aGF0IGlzIHJlcXVpcmVkLg0KDQpBcyBtdWNoIGFzIHlvdSBj YW4gYWZmb3JkLiAgU0NTSSBkaXNrIGFyZSBnZW5lcmFsbHkgYmV0dGVyLCB0 aG91Z2gNCm1vcmUgZXhwZW5zaXZlLg0KIA0KPiBpdikgUHJvYmxlbXMgd2l0 aCBmb3J0cmFuIGNvbXBpbGF0aW9uIG9uIHRoZXNlIFBDcywgYW5kIHdoYXQg Zm9ydHJhbg0KPiBjb21waWxlciBpcyBwcmVmZXJyZWQuDQoNClRoZSBvbmx5 IHN0YWJsZSBmcmVlIGZvcnRyYW4gJ2NvbXBpbGVyJyBvbiBsaW51eCBpcyBm MmMgY29udmVydGVyIGFuZA0Kc3JpcHRzIGJ1aWx0IGFyb3VuZCBpdC4gIEdO VSdzIGc3NyBwcmVzZW50bHkgaXMgaW4gYmV0YSBzdGFnZSBhbmQNCnRoZXJl IGFyZSB2YXJpb3VzIGVycm9ycyByZXBvcnRlZC4gIEknbSBub3QgZXhwZXJp ZW5jZWQgd2l0aA0KY29tbWVyY2lhbCBjb21waWxlcnMuDQogDQo+IHYpIEFu eSBvdGhlciBtYWNoaW5lcyB0aGF0IG1heSBiZSBiZXR0ZXIgc3VpdGVkIGZv ciBvdXIgYXBwbGljYXRpb25zLg0KDQpXZWxsLCBpZiB5b3UgY2FuIGFmZm9y ZCBEaWdpdGFsJ3MgQWxwaGEgKG5vdCBjbG9uZXMpIHVuZGVyIERpZ2l0YWwN ClVuaXggb3IgU2lsaWNvbiBSMTBLIG1hY2hpbmVzIC4uLg0KIA0KPiBJIHdp bGwgc3VtbWFyaXNlIHRoZSByZXNwb25zZXMsDQo+IA0KPiBNYW55IHRoYW5r cw0KPiBLaW0NCg0KU1ksDQpBbmRyZXkgViBLaGF2cnl1dGNoZW5rbyAgICAg ICAgICAgICAgICAgYWtoYXZyQGNvbXBjaGVtLmtpZXYudWENClN0ZC4gRGlz Y2xhaW1lcjogTXkgb3BpbmlvbnMgb25seS4NClRoZSB3aG9sZSBpZGVhIG9m IE9PUDogU2hhcmUgYW5kIFJldXNlLiAgU3VwcG9ydCB0aGUgRlJFRSBzb2Z0 d2FyZSENCg0KLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tDQpJ IHVzZSBQQ3MgZm9yIGNvbXB1dGF0aW9uYWwgd29yayBhdCB0aGUgbW9tZW50 LCBzbyBteSBwb2ludCBvZiB2aWV3IG1pZ2h0DQpiZSBoZWxwZnVsIGZvciB5 b3UuDQoNCg0KPiBpKSBUaGUgbW9kZWwgb2YgUGVudGl1bSBQQyB0aGF0IHdl IG5lZWQsIG9yIHdvdWxkIHByZWZlciwgdG8gcnVuIHRoZQ0KPiBhYm92ZSBw YWNrYWdlcywgZXNwZWNpYWxseSBHYXVzc2lhbiAoZS5nLiwgUGVudGl1bSBJ SSwgUGVudGl1bSBQcm8pLg0KDQpHbyBmb3IgYSBQUHJvIG9yIFBlbnRpdW0g SUkgaWYgeW91IGhhdmUgdGhlIG1vbmV5LiBBbmQgdHJ5IHRvIGdldCBob2xk IG9uDQphbGwgUGVudGl1bXMgeW91IGNhbiBnZXQuIFNvbWV0aW1lIGFnbyBJ IGdvdCBhIHVzZWQgMnhQZW50aXVtIGZyb20gYW4gb2xkIA0KZ3JhZCBzdHVk ZW50IHdobyBoYXMganVzdCBncmFkdWF0ZWQuIEV2ZXJ5Ym9keSB3YW50IExB UkdFIGNvbXB1dGVycyBzbw0KaXQncyBlYXN5IHRvIGdldCB0aGUgdXNlZCBQ ZW50aXVtcyBmb3IgYWxtb3N0IG5vdGhpbmcgOy0pLiBUaGUgZGVhbCBmb3IN CnlvdSBpcyB0aGF0IHlvdSBnZXQgbW9yZSBjb21wdXRpbmcgcG93ZXIgZm9y IGFsbW9zdCBub3RoaW5nLg0KDQpPbmUgcG9pbnQgdG8gY29uc2lkZXIgaXMg aGFyZGRpc2suIEkgaGF2ZSB0aHJlZSBtYWNoaW5lcyB3aXRoIEVJREUgZGlz a3MsDQpidXQgSSB0aGluayB0aGF0IFNDU0kgZGlza3MgbXVzdCBiZSBjb25z aWRlcmVkIGJlY2F1c2Ugb2Z0ZW4gdGhleSBhcmUNCmZhc3Rlci4gQWJvdXQg NSBHQiBwZXIgbWFjaGluZSBpcyBhIG1pbmltdW0uDQoNCj4gaWkpIElzIHRo ZXJlIGlzIGEgZGlzdGluY3QgYWR2YW50YWdlIGluIGhhdmluZyB0aGUgZHVh bCBDUFUgbWFjaGluZSwNCj4gb3IgaXMgaXQgYmV0dGVyIHRvIHN0aWNrIHdp dGggYSBzaW5nbGUgcHJlY2Vzc29yIFBDLg0KDQpJIGhhdmUgdHdvIGR1YWwg bWFjaGluZXMgKFBQcm9zIGFuZCBQZW50aXVtcykuIFRoZSBhZHZhbnRhZ2Ug aXMgdGhhdCB5b3UNCm9ubHkgaGF2ZSB0byBhZG1pbmlzdHJhdGUgb25lIGNv bXB1dGVyIHBlciB0d28gY3B1cyBhbmQgbm90IHR3byBjb21wdXRlcnMuDQpS dW5uaW5nIExpbnV4LCBJIGhhdmVuJ3QgZW5jb3VudGVyZWQgYW55IHByb2Js ZW1zIGFuZCBteSBmZWVsaW5nIGlzIHRoYXQNCnlvdSdyZSBvbmx5IGxvb3Np bmcgYWJvdXQgMC4xICUgcGVyIGNwdS4gSW4gb3RoZXIgd29yZHMsIGR1YWwg bWFjaGluZXMgaXMNCm15IGNob2ljZS4gTWFjaGluZXMgY29udGFpbmluZyBm b3VyIGNwdXMgYXJlIHRvbyBleHBlbnNpdmUgZm9yIG1lLg0KDQo+IGlpaSkg VGhlIGFtb3VudCBvZiBSQU0gYW5kIGhhcmQgZGlzayB0aGF0IGlzIHJlcXVp cmVkLg0KDQpJIHdvdWxkIGdvIGZvciA2NCBtZWdzIHBlciBjcHUuIA0KDQo+ IGl2KSBQcm9ibGVtcyB3aXRoIGZvcnRyYW4gY29tcGlsYXRpb24gb24gdGhl c2UgUENzLCBhbmQgd2hhdCBmb3J0cmFuDQo+IGNvbXBpbGVyIGlzIHByZWZl cnJlZC4NCg0KSSB1c2UgZWl0aGVyIGYyYyBvciBnNzcuIFNvbWV0aW1lcyBn NzcgZ2l2ZXMgbWUgcHJvYmxlbXMsIGJ1dCB0aGVuIEkgc2hpZnQNCnRvIGYy YyB3aGljaCBjYW4gY29tcGlsZSBldmVyeXRoaW5nIChhcyBsb25nIGFzIGl0 IGlzIHNob3J0IC0gYmFkDQpzZW50ZW5jZSwgc29ycnkpLiBUaGVyZSBpcyBh IGdjYyBmb3IgUGVudGl1bSBhbmQgUFBybyBidXQgaXQgZG9lcyBub3Qgc2Vl bQ0KdG8gYmUgc3RhYmxlLiBCb3RoIGYyYyBhbmQgZzc3IGNhbiB1c2UgaXQg KGc3NyB1c2VzIHRoZSBiYWNrZW5kIG9mIGdjYw0KaS5lLiB0aGUgY29kZSBn ZW5lcmF0b3IpLiBZZXN0ZXJkYXkgSSBmb3VuZCBCTEFTIGZvciBQZW50aXVt ICh3cml0dGVuIGluDQphc3NlbWJsZXIhKS4NCg0KDQpCZXN0IHJlZ2FyZHMN CiAgS2VubmV0aCBHZWlzc2hpcnQNCg0KDQorLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLSsN CnwgS2VubmV0aCBHZWlzc2hpcnQgICAgICAgICAgICAgICAgICAgICAgICAg ICAgIFBoLkQuLXN0dWRlbnQgfA0KfCBEZXB0LiBvZiBMaWZlIFNjaWVuY2Vz IGFuZCBDaGVtaXN0cnkgICAgUm9za2lsZGUgVW5pdmVyc2l0eSB8DQorLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLSsNCnwgICAgICAgICAgICAgICAgICAgICAgVml2YSBs YSBwZW5ndWluICAgICAgICAgICAgICAgICAgICAgICAgfA0KKy0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0rDQoNCg0KPiBpKSBUaGUgbW9kZWwgb2YgUGVudGl1bSBQQyB0 aGF0IHdlIG5lZWQsIG9yIHdvdWxkIHByZWZlciwgdG8gcnVuIHRoZQ0KPiBh Ym92ZSBwYWNrYWdlcywgZXNwZWNpYWxseSBHYXVzc2lhbiAoZS5nLiwgUGVu dGl1bSBJSSwgUGVudGl1bSBQcm8pLg0KDQpUaGVyZSdzIGxpdHRsZSBkaWZm ZXJlbmNlIGJldHdlZW4gdGhlIFBlbnRpdW1Qcm8gYW5kIHRoZSBQZW50aXVt IElJIGZvcg0KbnVtYmVyIGNydW5jaGluZywgc28gcGljayB3aGF0J3MgY2hl YXBlciBmb3IgeW91Lg0KDQo+IGlpKSBJcyB0aGVyZSBpcyBhIGRpc3RpbmN0 IGFkdmFudGFnZSBpbiBoYXZpbmcgdGhlIGR1YWwgQ1BVIG1hY2hpbmUsDQo+ IG9yIGlzIGl0IGJldHRlciB0byBzdGljayB3aXRoIGEgc2luZ2xlIHByZWNl c3NvciBQQy4NCg0KV2VsbCwgeW91IGdldCB0d2ljZSB0aGUgcHJvY2Vzc2lu ZyBwb3dlciwgYWx0aG91Z2ggc2hhcmVkIG1lbW9yeQ0KYWNjZXNzIG1lYW5z IHRoYXQgeW91IHdpbGwgbm90IHF1aXRlIGdldCB0aGUgc3BlZWQgb2YgdHdv IHNlcGFyYXRlDQptYWNoaW5lcy4NCg0KPiBpaWkpIFRoZSBhbW91bnQgb2Yg UkFNIGFuZCBoYXJkIGRpc2sgdGhhdCBpcyByZXF1aXJlZC4NCg0KVGhlIG1v cmUgdGhlIGJldHRlciwgb2YgY291cnNlLiBJIHdvdWxkbid0IHN0YXJ0IGJl bG93IDY0IE1CIG9mDQptYWluIG1lbW9yeS4gVGhlIHJlc3QgZGVwZW5kcyBv biB0aGUgY2FsY3VsYXRpb25zIHlvdSBwbGFuIHRvIHJ1bi4NCkhhcmQgZGlz ayBzaXplIGRlcGVuZHMgZW50aXJlbHkgb24geW91ciBkYXRhIHN0b3JhZ2Ug bmVlZHMuDQpUaHMgc3lzdGVtIGl0c2VsZiB3aWxsIGZpdCBoYXBwaWx5IHdp dGhpbiA1MDAgTUIsIGV2ZW4gd2l0aA0KcGxlbnR5IG9mICJsdXh1cnkiIHBy b2dyYW1zIGluc3RhbGxlZC4gQmUgZ2VuZXJvdXMgd2l0aCBzd2FwIHNwYWNl LA0Kc2luY2UgeW91IG1pZ2h0IHdhbnQgdG8gZXh0ZW5kIFJBTSBsYXRlci4N Cg0KPiBpdikgUHJvYmxlbXMgd2l0aCBmb3J0cmFuIGNvbXBpbGF0aW9uIG9u IHRoZXNlIFBDcywgYW5kIHdoYXQgZm9ydHJhbg0KPiBjb21waWxlciBpcyBw cmVmZXJyZWQuDQoNCk15IG9ubHkgZXhwZXJpZW5jZSBpcyB3aXRoIGc3Nywg d2hpY2ggY29tcGlsZXMgYWxsIGNvZGUgSSBoYXZlIHRlc3RlZA0Kd2VsbCBl bm91Z2guIEl0cyBlcnJvciBtZXNzYWdlcyBjYW4gYmUgY29uZnVzaW5nLCBz byBpdCdzIHBlcmhhcHMgbm90DQp0aGUgYmVzdCBjaG9pY2UgZm9yIGRldmVs b3BtZW50LiBJbiB0ZXJtcyBvZiBjb2RlIHF1YWxpdHkgaXQgc2hvdWxkDQp0 aGVvcmV0aWNhbGx5IGJlIHdvcnNlIHRoYW4gb3B0aW11bSwgYnV0IEkgaGF2 ZSBubyBudW1iZXJzIGF0IGhhbmQuDQoNCj4gdikgQW55IG90aGVyIG1hY2hp bmVzIHRoYXQgbWF5IGJlIGJldHRlciBzdWl0ZWQgZm9yIG91ciBhcHBsaWNh dGlvbnMuDQoNCkluIHRlcm1zIG9mIHByaWNlL3BlcmZvcm1hbmNlIHJhdGlv IFBDcyBhcmUgaGFyZCB0byBiZWF0LCBidXQgb2YNCmNvdXJzZSB5b3UgY2Fu IGdldCBoaWdoZXIgcGVyZm9ybWFuY2UgZnJvbSBvdGhlciBtYWNoaW5lcy4N Ci0tIA0KLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLQ0KS29u cmFkIEhpbnNlbiAgICAgICAgICAgICAgICAgICAgICAgICAgfCBFLU1haWw6 IGhpbnNlbkBpYnMuaWJzLmZyDQpMYWJvcmF0b2lyZSBkZSBEeW5hbWlxdWUg TW9sZWN1bGFpcmUgICB8IFRlbC46ICszMy00Ljc2Ljg4Ljk5LjI4DQpJbnN0 aXR1dCBkZSBCaW9sb2dpZSBTdHJ1Y3R1cmFsZSAgICAgICB8IEZheDogICsz My00Ljc2Ljg4LjU0Ljk0DQo0MSwgYXYuIGRlcyBNYXJ0eXJzICAgICAgICAg ICAgICAgICAgICB8IERldXRzY2gvRXNwZXJhbnRvL0VuZ2xpc2gvDQozODAy NyBHcmVub2JsZSBDZWRleCAxLCBGcmFuY2UgICAgICAgICB8IE5lZGVybGFu ZHMvRnJhbmNhaXMNCi0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0NCg0KDQoNCg0KDQo= ---1915151607-2078917053-875136674=:15959-- From the web I can see that Cornell goes on emphasizing KSR & IBM, Status: RO Pittsburgh and San Diego move on with Crays, and NCSA... I have the following specific questions: 1) Overall work load of the centers? It is sad that you get a job to run in several hours but after waiting in a queue for several days. I am interested in actual wall clock time for the turn around of jobs. 2) Any center with special interest or resources on molecular dynamics? 3) Do centers prefer that users make upgrades to their local resources first, even if these cost several $10K, before granting time? 4) Are there any centers with special considerations for initial applicants (although not first time supercomputer users) with pilot projects? Thanks in advance for your replies. -- ("`-/")_.-'"``-._ Gustavo A. Mercier,Jr.,MD,PhD (. . `) -._ )-;-,_() Division of Nuclear Medicine (v_,)' _ )`-.\ ``- Dept. of Radiology _;- _,-_/ / ((,' University of Pennsylvania ((,.-' ((,/ 3400 Spruce St. gmercier@mail.med.upenn.edu Philadelphia, PA 19104 215-662-3069/3091 fax: 215-349-5843 From tkwoo@acs.ucalgary.ca Fri Sep 26 19:26:50 1997 Received: from ds11.acs.ucalgary.ca for tkwoo@acs.ucalgary.ca by www.ccl.net (8.8.3/950822.1) id TAA23437; Fri, 26 Sep 1997 19:09:08 -0400 (EDT) Received: by acs5.acs.ucalgary.ca (AIX 4.1/UCB 5.64/4.03) id AA89804; Fri, 26 Sep 1997 17:04:40 -0600 Message-Id: <9709262304.AA89804@acs5.acs.ucalgary.ca> Subject: SCSI needed on Pentium Box for DFT? To: CHEMISTRY@www.ccl.net Date: Fri, 26 Sep 97 17:04:39 MDT From: "Tom K. Woo" X-Mailer: ELM [version 2.3 PL11K] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Greetings all, We are planning to buy some Intel Pentium II or Pentium Pro machines to run some DFT calculations on (ADF to be exact). I would like to generate some discussion on the necessity of using SCSI disks as opposed to EIDI or ultraDMA(PII only) disks for running these types of codes. The reason I ask this is that SCSI disks are about twice the price of non-SCSI disks. If a SCSI disk is not required, then the money saved can be used to upgrade from a 233 Mhz Pentium II to a 300 Mhz Pentium II. I understand that with EIDE or ultra DMA it is the Pentium that actually controls the disk, while with SCSI disks has its own controller. Anybody have a feeling for how this effects the performance of a typical DFT code?? (I have a pentium at home with an EIDE drive, and it is as fast or faster than any SCSI disks I have had experience with.) Thanks in Advance. Tom Woo ******************************************************************************** * Tom Woo voice:(403) 220-7176 * * Graduate Student fax: (403) 289-9488 * * Department of Chemistry email: tkwoo@zinc.chem.ucalgary.ca * * University of Calgary homepage: http://www.ucalgary.ca/~tkwoo * * 2500 University Drive, N.W. room: SB 427 * * CALGARY, Alberta, Canada, T2N 1N4 * ******************************************************************************** From dsf@world.std.com Sat Sep 27 09:26:57 1997 Received: from europe.std.com for dsf@world.std.com by www.ccl.net (8.8.3/950822.1) id JAA25137; Sat, 27 Sep 1997 09:09:31 -0400 (EDT) Received: from world.std.com by europe.std.com (8.7.6/BZS-8-1.0) id JAA01471; Sat, 27 Sep 1997 09:09:31 -0400 (EDT) Received: from world.std.com by world.std.com (5.65c/Spike-2.0) id AA18333; Sat, 27 Sep 1997 09:09:27 -0400 Message-Id: <2.2.16.19970927130705.26e72fcc@world.std.com> X-Sender: dsf@world.std.com (Unverified) X-Mailer: Windows Eudora Pro Version 2.2 (16) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Sat, 27 Sep 1997 09:07:05 -0400 To: chemistry@www.ccl.net From: Dawn Friedman Subject: G:G94 RIC and GAMESS: constrain var to equivalence? This is only partly a G94 question -- my real concern is with GAMESS. I'm geometry-optimizing some model systems whose high symmetry is broken only by methyl rotors. I want the methyl hydrogens to be free to get out of each other's way, but I want the C-C frame bonds to stay equivalent, at least until I have a partial optimization. This is easy to do in a Gaussian z-matrix format. G94 can be forced to use the input zmatrix rather than redundant internal coordinates, though of course all the advantages of the RICs are lost. (Is there a mathematical reason why G94 can't construct an RIC set with some variables constrained to be equal?) I haven't figured out *any* way to constrain a set of bond lengths or angles to be equal *but variable* in GAMESS when the entire molecule doesn't have symmetry. My choices seem to be: (a) Force the entire molecule, including all the methyl hydrogens, to exhibit at least one symmetry element. (b) Add fixed constraints in $STATPT, which will hold the C-C bond lengths to a constant value, and hope I picked the right one :-) I'm working with model systems that aren't necessarily stable as single molecules in a full optimization, so I can't count on their optimizing to an energy minimum if I don't constrain them. Should I just choose option (a) and let the CH3 hydrogens suffer? Is the hybrid I'm trying to create -- symmetric frame and free methyl hydrogens -- inherently invalid? I'll summarize to the list. Thanks, Dawn Friedman Dawn Friedman dsf@world.std.com From taisung@chem.duke.edu Sat Sep 27 22:27:04 1997 Received: from chem.duke.edu for taisung@chem.duke.edu by www.ccl.net (8.8.3/950822.1) id VAA26652; Sat, 27 Sep 1997 21:56:14 -0400 (EDT) Received: from debye.chem.duke.edu (debye.chem.duke.edu [152.3.169.112]) by chem.duke.edu (8.7.4/Duke-1.1) with ESMTP id VAA28015; Sat, 27 Sep 1997 21:55:58 -0400 (EDT) Received: (taisung@localhost) by debye.chem.duke.edu (8.7.4/Duke-1.1) id VAA15678; Sat, 27 Sep 1997 21:55:56 -0400 (EDT) Date: Sat, 27 Sep 1997 21:55:53 -0400 (EDT) From: Taisung Lee X-Sender: taisung@debye To: CAVALLO@CHEMNA.DICHI.UNINA.IT cc: CHEMISTRY@www.ccl.net Subject: Re: CCL:PCI vs. SCSI and Linux In-Reply-To: <01IO4X524BR40001H3@CHEMNA.DICHI.UNINA.IT> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear CCLers, There is a little misunstanding in Cavallo's article. A PCI bus is a bus between CPU and the world outside it. A SCSI interface is the bridge between the system bus and your HD or other devices. You can not compare these two just as you can not say "My CPU is faster than your hard disk." Although this issue is not directly related to computational chemistry, PC's can be much cheaper choices to us CCLers. We shall to know little more about PC. A PC internal structure looks like: CPU<---L2 Cache<---Main Memory | | System Bus------------------------------ | | | | | | | | Video Keyboard SCSI IDE interface Interface , mouse.... interface | | | | Video device SCSI IDE device device To let your PC fly, the first thing to do is get a fast CPU. To do computations, Pentium Pro-200 or Pentium II's will be good choice. The second thing is get as much memory as you can. Give the current price ~$3.5 /MB, ther is no reason to cut your memory. For other parts of the system, PCI is one kind of System bus. Curreny PCI standard has highest possible bandwith of 133 MB/s in 32 bit mode and 266 MB/s in 64 bit mode. Almost new PCs come with PCI bus otherwise it is too old to keep or it is too new... We probably don't have other choices now. For people doing visualization, a good video card is necessary. Remember only get a PCI version video card. You may see 'ISA' version. ISA is old standard of system bus which has the bandwith of 8 MB/s. Most PC's keep some ISA slots. But don't buy an ISA video card or you will feel your Pentium PC is slower than a 486 when you run Rasmol. As Luigi Cavallo said, HD is another important part to look at. The most popular HDs are IDE (or its variation, EIDE and Ultra-IDE, ATA-2, ATA-3,..) HD and SCSI HD. SCSI HDs used to be much faster than IDEs. But it is not the case anymore. However, SCSI still has some advantages. We need to understand what are the factors to affect the HD performace. The first factor, of course, is the spinning speed of the HD. Most HDs has the spinning speed ~5400 RPM. High end SCSI HDs have 7200 RPM. Seagate has the fastest HD in the world with 10000 RPM. (Don't expect your workstation has better HDs than a new PC...) HDs have buffer memory to store data when do reading/writing. So the other factors will be the size of buffer memory and the data transfer rate between disks and buffer (called internal transfer rate). You sholud get a HD with buffer no less than 256KB (except IBM's, they can make very fast HD with 128KB buffer and nobody knows why...) and internal transfer rate no less than 100 Mbit/s. High spinning HDs are all SCSI (>7200RPM) but not all SCSI HDs are high spinning. Be careful about tis point. You can buy an Ultra-Wide SCSI HD with only 4500RPM ( and for sure it is slow). SCSI and IDE HD have similar buffer size and internal transfer rate. The last factor (notice: I say it is the LAST factor) is the data transfer rate between HD buffer and the system bus. It is important only when you have very slow interface or your have multiple HDs. Following is a list of maximum transfer of different interfaces: IDE: normal IDE: 5MB/s (?) ATA-2, EIDE (mode 4): 16.6 MB/s ATA-3, UIDE : 33.2 MB/s SCSI: SCSI :2.5 MB/s SCSI-2:5 MB/s fast-SCSI-2:10MB/s wide-SCSI-2:10MB/s fast,wide-SCSI-2:20MB/s ultra,wide-SCSI:40MB/s Like chemical reactions, the HD performance has a 'rate-determining' step. Say you have a HD with internal transfer rate 120Mbit/s=15 MB/s (which is very good) and you get an UW-SCSI interface and PCI system bus (133MB/s). You will find your system is not faster at all compared with a similar HD with FW-SCSI interface. If you buy an 4500RPM, internal transfer rate=50Mbit/s HD and buy an UW-SCSI card, you may just waste your money. I don't want to write too much here. To be short, don't be fooled by the bandwith of IDE and SCSI. A good EIDE HD will be as faster as a FW-SCSI/UW-SCSI HD. The HD performace of a new PC is very likely better than the workstation in your lab. However, if you want to get very very good performance, SCSI is the only way. The best HDs are still SCSI (with a doubled price). A well-develped technique called RAID allows system to write/read data to/from several SCSI HDs at the same time. Using RAID you can reach 150MB/s easily ( at that time, the PCI bus becomes the bottleneck). And you can expect 10 times more money will be gone... I already wrote too much. If there is any interest on this topic, I will post more about it. Such as the sync. transfer and multiple commands of SCSI bus. My suggestion on HD for PC. Get a new 5400RPM EIDE one and you will be happy. If you can spend at least as twice money, get a high end SCSI. my $0.02 opinion Taisung Lee > Dear CCLers, > > Recent discussions on the list "seem" to assess that the PPro 256kb > should be the best buy, in terms of performance/costs ratio. > > Anyhow, the PCs are not simply made by the CPU. We know that another > bottleneck can be the HD. A PCI bus should be more or less comparable > to a FAST SCSI 2. Better performances can be obtained on going to > Ultra Wide etc. SCSI stuff. Anyhow, prices increase. > > What the experience here ? > > Are those (more) expensive last generation SCSI pieces really needed ? > > Any comments appreciated > > Best regards, > > Luigi > > ----------------------------------- > Luigi Cavallo > Dept. of Chemistry > Univ. of Naples > Email cavallo@chemna.dichi.unina.it > ----------------------------------- > > -------This is added Automatically by the Software-------- > -- Original Sender Envelope Address: CAVALLO@CHEMNA.DICHI.UNINA.IT > -- Original Sender From: Address: CAVALLO@CHEMNA.DICHI.UNINA.IT > CHEMISTRY@www.ccl.net: Everybody | CHEMISTRY-REQUEST@www.ccl.net: Coordinator > MAILSERV@www.ccl.net: HELP CHEMISTRY or HELP SEARCH | Gopher: www.ccl.net 73 > Anon. ftp: www.ccl.net | CHEMISTRY-SEARCH@www.ccl.net -- archive search > Web: http://www.ccl.net/chemistry.html > >