From bauerdick@fnal.gov Fri Apr 10 22:28:41 2009 Return-Path: X-Original-To: harvey@monalisa.cern.ch Delivered-To: harvey@monalisa.cern.ch Received: from cernmx13.cern.ch (cernmx13.cern.ch [137.138.142.200]) by monalisa.cern.ch (Postfix) with ESMTPS id 911A128020152 for ; Fri, 10 Apr 2009 22:28:41 +0200 (CEST) Received: from neutrino.hep.caltech.edu (131.215.116.43) by cernmxgwlb.cern.ch (137.138.142.186) with Microsoft SMTP Server (TLS) id 8.1.340.0; Fri, 10 Apr 2009 22:28:40 +0200 Received: from localhost ([127.0.0.1]) by neutrino.hep.caltech.edu with esmtp (Exim 4.63) (envelope-from ) id 1LsNL3-0005N3-61 for newman@hep.caltech.edu; Fri, 10 Apr 2009 13:28:39 -0700 X-Virus-Scanned: amavisd-new at hep.caltech.edu Received: from neutrino.hep.caltech.edu ([127.0.0.1]) by localhost (mail.hep.caltech.edu [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Gh9E9k67Qqsr for ; Fri, 10 Apr 2009 13:28:33 -0700 (PDT) Received: from mailgw2.fnal.gov ([131.225.111.12]) by neutrino.hep.caltech.edu with esmtp (Exim 4.63) (envelope-from ) id 1LsNKx-0005Mx-G8 for newman@hep.caltech.edu; Fri, 10 Apr 2009 13:28:33 -0700 Received: from mailav1.fnal.gov (mailav1.fnal.gov [131.225.111.18]) by mailgw2.fnal.gov (iPlanet Messaging Server 5.2 HotFix 2.06 (built Mar 28 2005)) with SMTP id <0KHW005O8JGMHE@mailgw2.fnal.gov> for newman@hep.caltech.edu; Fri, 10 Apr 2009 15:27:24 -0500 (CDT) Received: from mailgw2.fnal.gov ([131.225.111.12]) by mailav1.fnal.gov (SAVSMTP 3.1.7.47) with SMTP id M2009041015272323997 for ; Fri, 10 Apr 2009 15:27:23 -0500 Received: from conversion-daemon.mailgw2.fnal.gov by mailgw2.fnal.gov (iPlanet Messaging Server 5.2 HotFix 2.06 (built Mar 28 2005)) id <0KHW00L01IXSVR@mailgw2.fnal.gov> (original mail from bauerdick@fnal.gov) for newman@hep.caltech.edu; Fri, 10 Apr 2009 15:27:23 -0500 (CDT) Received: from [10.0.1.121] (c-71-194-100-25.hsd1.il.comcast.net [71.194.100.25]) by mailgw2.fnal.gov (iPlanet Messaging Server 5.2 HotFix 2.06 (built Mar 28 2005)) with ESMTPSA id <0KHW0053KJHMHF@mailgw2.fnal.gov>; Fri, 10 Apr 2009 15:27:23 -0500 (CDT) Date: Fri, 10 Apr 2009 15:27:21 -0500 From: LATBauerdick Subject: letter of support for US LHCnet Supplemental To: Harvey B Newman , Artur Barczyk CC: Joel Butler , Ian Fisk , Daniel Marlow Message-ID: MIME-Version: 1.0 X-Mailer: Apple Mail (2.930.3) Content-Type: multipart/signed; protocol="application/pkcs7-signature"; micalg=sha1; boundary="Apple-Mail-1--904962891" X-Spam-Score: -6.6 (------) Received-SPF: Neutral (cernmxgwlb.cern.ch: 131.215.116.43 is neither permitted nor denied by domain of bauerdick@fnal.gov) X-CERN-Reverse-DNS: neutrino.hep.caltech.edu Keywords: CERN SpamKiller Note: 1 --Apple-Mail-1--904962891 Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Dear Artur, dear Harvey, This letter is to support the supplemental funding request for US LHCNet circuits in 2009. This supplemental request is vital for US CMS and US ATLAS to get the required Transatlantic network support. We believe that the proposed solution is cost optimized while meeting the technical requirements for path diversity and vendor reliability. The request outlines the operational requirements for US LHCnet, which result from the bandwidth needs and the needs for sustained and reliable operations of the transatlantic network between CERN and the U.S. Tier-1 centers. For the CMS experiment the US LHCnet can be seen as an extension of the Data Acquisition network, given the crucial role that the Fermilab Tier-1 center plays for the CMS data taking and data analysis. Accordingly, there are important requirements on reliability and availability of this network. Any loss of connectivity and even loss of available bandwidth will have important impact on our ability to take good quality data and to make it available to the US physics community for data analysis and discovery. In the CMS computing model there is only one active copy of CMS raw data, which is at one of the Tier-1 centers, with Fermilab being by far the largest at some 40% of the total available Tier-1 resources. The CERN computing center does not provide enough throughput to access the second archive copy at CERN while data is being taken and processed at the CERN Tier-0. The services hosted at Fermilab are essential for the well functioning of the CMS data grid system. Should in extreme cases the transatlantic network fail and the buffers at CERN flow over, that data would only be archived, but not become available for data processing, likely until the end of the data taking period. The US LHCNet is also used for bringing the re-processed datasets into circulation for analysis. This data flow will for some time use a significant part of the provided bandwidth. We expect that in particular during the first years of operations there will be rapid development of detector understanding and calibration, requiring successive re-processing of data at the Tier-1 centers. The US LHCnet both provides the US access to the majority of reprocessed data not hosted in the US, and the rest of the experiment access to data reprocessed at Fermilab. The well functioning of this data and workflow is essential for the experiment to extract the early physics and discoveries. The provisioning of redundant links thus seems prudent and necessary, and the US LHCnet plan foresees sufficient bandwidth to allow expedient and efficient analysis of CMS data. In particular it is vital for the U.S. participants in CMS to access data for physics analysis and discovery. There is a history of link-related network outages, leading to reductions in bandwidth for many hours, and in some cases lost connectivities on these vital life lines for the experiments. With the huge role that the Fermilab Tier-1 center plays for CMS computing, and with the essential role that getting the CMS data into our Tier-1 center has for the US physics efforts, this would be unacceptable in a real data taking scenario. The plan in the proposed supplement addresses these issues and foresees the circuit multiplicity and bandwidth at a level just sufficient to cover the specific needs for the US communities and our Tier-1 centers. The US LHCnet roadmap has so far been very successful to fulfill the transatlantic networking needs for the CERN-Tier-1 connections. We view this supplement request vital for the continued success of the US LHCnet and thus the CMS and LHC physics program in the United States. Sincerely, Lothar Bauerdick, US CMS S&C Manager concurring: Joel Butler, U.S. CMS Operations Program Manager Daniel Marlow, U.S. CMS Deputy Research Program Manager --Apple-Mail-1--904962891 Content-Disposition: attachment; filename="smime.p7s" Content-Type: application/pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIECDCCBAQw ggLsoAMCAQICAncnMA0GCSqGSIb3DQEBBQUAMGkxEzARBgoJkiaJk/IsZAEZFgNvcmcxGDAWBgoJ kiaJk/IsZAEZFghET0VHcmlkczEgMB4GA1UECxMXQ2VydGlmaWNhdGUgQXV0aG9yaXRpZXMxFjAU BgNVBAMTDURPRUdyaWRzIENBIDEwHhcNMDkwMjA2MTkzMzE4WhcNMTAwMjA2MTkzMzE4WjBiMRMw EQYKCZImiZPyLGQBGRYDb3JnMRgwFgYKCZImiZPyLGQBGRYIZG9lZ3JpZHMxDzANBgNVBAsTBlBl b3BsZTEgMB4GA1UEAxMXTG90aGFyIEJhdWVyZGljayAyMTIxMjMwggEiMA0GCSqGSIb3DQEBAQUA A4IBDwAwggEKAoIBAQD82164ZFSWG9J2FMt1CAbqiSDjdpMZbcts3hzE7PoIq9GXL3qS1d45CZmk H028Daa8k1QMwH0iV2l7JeqFF0iaKz+ll5EzDYVb26UdMOGrUZwX7X2BxQ/qefDxJfJlr9V6JFe+ BMtPpPgdstKRwudzaw5SKxnbw3k0W7G/I6zfD3oHf7SPviG5yC3H7pjyEy+CT746eQUUiP3J7o8B 0pFltI5H1zxRWdPb4atm2vmtSPomViMRI6l2yE8brDeYnh3C1apwRIwmNVgTYFErNoMKpfkceEbv kjs2EAQMYQQhGyBHf6Uromkw4I9o9z4rFBayfLwd9SoqwJ1AJghe9XHTAgMBAAGjgbwwgbkwEQYJ YIZIAYb4QgEBBAQDAgWgMA4GA1UdDwEB/wQEAwIF4DAfBgNVHSMEGDAWgBTKGR0Sjm6kOF1C1DEO CNvZjRcNXTAYBgNVHSAEETAPMA0GCyqGSIb3TAMHAQIKMDoGA1UdHwQzMDEwL6AtoCuGKWh0dHA6 Ly9wa2kxLmRvZWdyaWRzLm9yZy9DUkwvMWMzZjJjYTguY3JsMB0GA1UdEQQWMBSBEmJhdWVyZGlj a0BmbmFsLmdvdjANBgkqhkiG9w0BAQUFAAOCAQEAgxDF0I09TeDbpnIqwVtUtr2xW4x4/PlzaRQN P0D2fYcMqh0sg8Vv/Hy4nPtpKp4WvyKNABvlkIPqEWQ43Vvsz9QdzXo5MlwuqV2KwYITfI/A//ki GkMNa/fHTma3aZb8YUhzdBTh0+L4LrvUYYwQssndm1jgA1P9GCwMMVEbT2VplZKwQiS4hOyHsdXx 5uluTZ0X0Z2b+SkyxFl26leB8I0fUa9LQH609zex0qQdrw2tciANCxZsttZRDSTsNlGkzi1IJQnB 06Qq+qCnBoM+yiXa0p2AovfUnMTReLtbjfIANyYr88BmtD+oiLpmM5ufIWshR0LbIkWwNRKrw/rp jDGCAvowggL2AgEBMG8waTETMBEGCgmSJomT8ixkARkWA29yZzEYMBYGCgmSJomT8ixkARkWCERP RUdyaWRzMSAwHgYDVQQLExdDZXJ0aWZpY2F0ZSBBdXRob3JpdGllczEWMBQGA1UEAxMNRE9FR3Jp ZHMgQ0EgMQICdycwCQYFKw4DAhoFAKCCAWAwGAYJKoZIhvcNAQkDMQsGCSqGSIb3DQEHATAcBgkq hkiG9w0BCQUxDxcNMDkwNDEwMjAyNzIyWjAjBgkqhkiG9w0BCQQxFgQUfyegebLgCBWjj0loF5F/ UCw8+38wfgYJKwYBBAGCNxAEMXEwbzBpMRMwEQYKCZImiZPyLGQBGRYDb3JnMRgwFgYKCZImiZPy LGQBGRYIRE9FR3JpZHMxIDAeBgNVBAsTF0NlcnRpZmljYXRlIEF1dGhvcml0aWVzMRYwFAYDVQQD Ew1ET0VHcmlkcyBDQSAxAgJ3JzCBgAYLKoZIhvcNAQkQAgsxcaBvMGkxEzARBgoJkiaJk/IsZAEZ FgNvcmcxGDAWBgoJkiaJk/IsZAEZFghET0VHcmlkczEgMB4GA1UECxMXQ2VydGlmaWNhdGUgQXV0 aG9yaXRpZXMxFjAUBgNVBAMTDURPRUdyaWRzIENBIDECAncnMA0GCSqGSIb3DQEBAQUABIIBAPHU AB9FuJgvmNCGekJkD9d2PmSmnKK9tC582BZ/D/0qTTzZ1ivnPkD+NVmw+0+GXG7FUj6PrY855B1V Zdpt1Ax+kSqULswQFzyGndKsU4Kamvrz870+e+gZjvole9mzXL0sDjw8bIDAi869uMXD+f/q32nS nd9Fr+hGy8d0ruxExfErxEQYFCkRaWappnKklFYt35i8c+q/dN3DQnj0dh8fYcOlY/H8dEbz/Rzo lu+L6lWUMqENsyzS06s75DRUWQbzvZHufXHLqV21xc7JaM7KesfHxZ3Erq/nJfBiH/Yz3ePlA66r ECgtFTI9xih/bZQLxlrqxUIsExQ4r1uOhz4AAAAAAAA= --Apple-Mail-1--904962891--