- ur-wg - lists.ogf.org

Next phone meeting will be today the 30th of January at 15:30 CET
by Andrea Cristofori 30 Jan '13

30 Jan '13

Dear all, I would like to remember you our phone meeting today at 15:30 CET. If can find an agreement in the issue discussed last week and explained in the last mail sent by Jon we can proceed with the proof reading and then the comment phase. Please note that in the latest version of the document, available here: https://redmine.ogf.org/dmsf_files/12994?download= we added SourceAddress and DestinationAddress attributes added in the NetworkInboundUsed and NetworkOutboundUsed respectively. Cheers, Andrea -- Andrea Cristofori INFN-CNAF Viale Berti Pichat 6/2 40127 Bologna Italy Tel. : +39-051-6092920 Skype: andrea-cnaf

1 1

ExecutionHost - flat or hierarchical
by Jon Kerr Nilsen 30 Jan '13

30 Jan '13

Hi, We had a discussion on the last phone meeting about the ExecutionHost in the ComputeUsageBlock (section 5.5 in http://redmine.ogf.org/dmsf_files/12990?download= ). At the beginning of writing up the UR-2.0 draft we decided that we should go for a flat structure of elements in the various usage blocks. While this has been a good approach for most of the elements, it has turned a bit tricky for the ExecutionHost. Constraints: - The ExecutionHost needs to support accounting for parallel jobs. - Parallel jobs means multiple hostnames. - Parallel jobs on multicore hosts means multiple process IDs per host - Multiple hosts means multiple benchmarks. - The process IDs and benchmarks are associated with specific hosts. - In XML you cannot have the same attribute multiple times in the same element. As far as I can see, there are two viable ways to meet these constraints. 1) What is already in the draft. ExecutionHost is a container for HostName, ProcessID and Benhmark. You can have multiple ExecutionBlocks per ComputeUsageBlock. You can have multiple ProcessIDs per ExecutionHost, and one HostName and one Benchmark. This meets all the constraints but means that the ComputeUsageBlock is not flat. 2) ExecutionHost is a string containing the hostname, and there are separate elements for ProcessID and Benchmark. You can have multiple ComputeUsageBlocks, one per host. There can only be one ExecutionHost per ComputeUsageBlock, but multiple ProcessIDs (all corresponding to the one ExecutionHost). The ComputeUsageBlocks can be identified (if identification is needed) as belonging to the same job e.g. by them having the same start time and, time wall time and cpu time. The Node element will always be 1 and should be removed. The number of processors used by the job can be found by summing the Processor elements in each of the This meets all the constraints and gives a flat structure in the ComputeUsageBlocks but means more work for parsers (and more room for mistakes) when accounting for parallel jobs. What do you think? Any opinions? cheers, Jon

1 0

Next phone meeting will be the 22nd of January at 15:00 CET
by Andrea Cristofori 22 Jan '13

22 Jan '13

Dear all, we updated the UR draft and at this link you can find the latest version: https://redmine.ogf.org/dmsf_files/12989?download= The most important updates are related to the cloud fields. There are some changes especially in the Attribute Matrix on the column of FedCloud. Now you can find CloudType in correspondence of HostType for the computing, memory and storage resources. The other change is related to SuspendTime that has been removed from FedCloud in favour of SuspendTime like in UR-2.0. We are also thinking to add a network block with some basic fields like in the FedCloud. What do you think? Anyone against it? We would also like to remind you that next phone meeting will be the 22nd of January at 15:00 CET. Cheers, Andrea and Jon -- Andrea Cristofori INFN-CNAF Viale Berti Pichat 6/2 40127 Bologna Italy Tel. : +39-051-6092920 Skype: andrea-cnaf

3 8

Wrap up of last phone meeting on 8th of January at 15:00 CET
by Andrea Cristofori 09 Jan '13

09 Jan '13

Dear all, After yesterday phone meeting we approved the changes proposed by Ralph from the beginning upt to MemoryUsageBlock excluded. The most important changes are the followings: - We removed the requirement MUST from the Infrastructure, Middleware and SubmitHost fields. - We also decided to move the description for ServiceLevel on the JobUsageBlock to ComputingUsageBlock as a part of the Host element and rename it as Benchmark. The Host element itself has been modified and it is now a container for HostName, Benchmark and ProcessId. In this way it can be used multiple Benchmarks and ProcessId connected to the same Host. - ServiceLevel in JobUsageBlock is now only a descriptive string We also discussed about if we should have the Charge fields or not because it is not really relevant for accounting as it is something related to billing. If we decide to keep it probably we should simplify the current definition by removing the formula and define as a single float. The meaning of this value is then up to the site. What do you think? The direct link to the latest PDF file is the following: https://redmine.ogf.org/dmsf_files/12985?download= We will continue to work on the document and send updates. Next phone meeting will be the 22nd of January at 15:00 CET. Cheers, Andrea and Jon

1 0

Next phone meeting on 8th of January at 15:00 CET
by Andrea Cristofori 07 Jan '13

07 Jan '13

Dear all, first of all happy new year! I would like to remind you our next phone meeting tomorrow the 8th of January at 15:00 CET. During the holidays Ralph most kindly provided several correction to our text. The phone meeting will mostly be used to discuss these corrections/questions. Kind regards, Andrea and Jon -- Andrea Cristofori INFN-CNAF Viale Berti Pichat 6/2 40127 Bologna Italy Tel. : +39-051-6092920 Skype: andrea-cnaf

1 0

New Draft available and next phone meeting
by Andrea Cristofori 04 Jan '13

04 Jan '13

Dear all, We updated the UR 2.0 draft and we uploaded all the files to redmine. The pdf files is available at the usual location. The direct link to latest pdf is the following: https://redmine.ogf.org/dmsf_files/12979?download= If you have some time during the holidays please have a look at it and send your comment. We think that, except for some examples missing we are at a good point with all the blocks. The only one that might need more discussion is the CloudUsageBlock. We are still interacting, especially with Alison, on its definition but also in this case we think we are quite close to the final definition. We could think a bit more if we should include or not a block for the network usage. We can discuss this right after the holidays and take a final decision on next phone meeting which is fixed for Tuesday the 8th of January 2013 at 15:00 CET. Merry Christmas an Happy new Year, Andrea and Jon -- Andrea Cristofori INFN-CNAF Viale Berti Pichat 6/2 40127 Bologna Italy Tel. : +39-051-6092920 Skype: andrea-cnaf

2 2

Next phone meeting on 11th of December at 15:00 CET
by Andrea Cristofori 11 Dec '12

11 Dec '12

Dear all, the 11th at 15:00CET of December we will have our last phone meeting before Christmas holydays. As we said last week we migrated the document to latex. Everything is now available on Redmine. The source can be found here: http://redmine.ogf.org/projects/ur-wg/repository/revisions/master/show/ur-2… Howto get access: http://redmine.ogf.org/projects/ur-wg/repository/revisions/master/entry/REA… and you can also download the last compiled pdf version from this link: http://redmine.ogf.org/dmsf_files/12975?download= Please have a look at it and suggest any changes. If someone has already written some example we can add them and discuss them together. We think that, except for the part related to the Cloud and Network bloks the other sections are now quite clear and they need to be checked for the language. If John Alan Kennedy is still willing to help us with this task he could start with the first 8 sections. If Alison has some news about Network and the attribute for Cloud she could report during the meeting for discussion. Please send us the following information so we will add them to the document in the contributors section: First Name and Last Name Institution1 Address Country Email: jdoe(a)example.com Cheers, Andrea and Jon

3 3

Switch from .doc to .tex?
by Andrea Cristofori 28 Nov '12

28 Nov '12

Dear all, As we are moving forward with the work on our document, we would like to ask you a question. The document itself needs to be completely reworked for what concerns the layout, formatting, etc. We are thinking to move to latex. This will allow us to edit the source with different text editor without problems and to split the section for editing by different person. We can also use a tracking system like git already provided on the OGF portal. An introduction to git is available here: http://githowto.com/ If there no one is against the switch we can proceed and give you, soon, more detail. Please cast your preference by Monday, the 3rd of December. Cheers, Andrea and Jon -- Andrea Cristofori INFN-CNAF Viale Berti Pichat 6/2 40127 Bologna Italy Tel. : +39-051-6092920 Skype: andrea-cnaf

1 0

FW: Next phone meeting on 12th of November at 15:00 CEST
by alison.packer＠stfc.ac.uk 27 Nov '12

27 Nov '12

Hi All, I just checked the data in the cloud accounting database we are running for the FedCloud testbed again as I thought we should have some records with network activity data. Although we do not have the "NetworkType" field used in any of them, we do have data for some records in the "NetworkInbound" and "NetworkOutbound" fields, but just from sites who are running OpenNebula not from sites with OpenStack. So, I think we should keep those 2 fields. Do you agree? I will raise the issue with the FedCloud resource providers regarding getting "NetworkType" information in the cloud records as well. Thanks, Alison -----Original Message----- From: Andrea Cristofori [mailto:andrea.cristofori@cnaf.infn.it] Sent: 14 November 2012 14:01 To: Packer, Alison (STFC,DL,SC) Subject: Re: Next phone meeting on 12th of November at 15:00 CEST Hi Alison, No problem. I just hope you are better now. Yesterday we had a look to the "Cloud Accounting Usage Record" attributes in your document and I have some question: - network proprieties (NetworkType, NetworkInbound, NetworkOutbound) how do you think to measure that? Have you already checked if some metrics are available on different cloud solutions? - How do you think to use ZoneName and TimeZone? TimeZone we believe is not really needed if in the other timestamps the value is set using for example ISO 8601. It would be implicit. ZoneName looks like a way to group different Site. The accounting portal does several grouping based on different needs (coutry, VO, etc). The grouping for ZoneName might be something that can be set at the portal level. - SuspendTime: what happens if for example a record spans from time T0 to time T10 and the virtual machine was suspended from T4 to T6? To avoid problem you should force to write a new record if the machine come back from the suspend. Instead of the moment in which the machine was suspended we thought that it might be better to use this attribute for the amount of time the machine was suspended in the interval T10 - T0. What do you think? Next phone meeting is fixed for Tuesday the 27th of November. If you can you are welcome to join the discussion. You can find the last update to our document here: https://redmine.ogf.org/dmsf_files/12966?download= Did you tried to register to the redmine web page? Cheers, Andrea On 11/14/2012 10:49 AM, alison.packer(a)stfc.ac.uk wrote: > Andrea, > I am very sorry I missed this meeting because I have been off sick. > > Thanks, > Alison > > -----Original Message----- > From: Andrea Cristofori [mailto:andrea.cristofori@cnaf.infn.it] > Sent: 12 November 2012 14:56 > To: ur-wg(a)ogf.org > Cc: Packer, Alison (STFC,DL,SC) > Subject: Next phone meeting on 12th of November at 15:00 CEST > > Dear all, > > I would like to remind you tomorrow phone meeting at 15:00 CEST. We are continuing the improvements to the UR definition that we might discuss during the phone. At this address you can find the latest version of our document: > > https://redmine.ogf.org/dmsf_files/12964?download= > > We made some clean up (more work is of course needed). > > Please also note the inclusion of a new block CloudUsageBlock which > includes some attributes that we could use for cloud accounting. The > same attributes and a comparison with the attributes present in the > "Cloud Accounting Usage Record Schema" definition > (https://wiki.egi.eu/wiki/Fedcloud-tf:WorkGroups:Scenario4 provided by Alison Packer in CC) is presented in the Attribute Matrix. > The only attributes that have no correspondence at this time are those dealing with network usage and TimeZone and ZoneName. > If you can not be present to tomorrow phone meeting please have a look in advance to the document and send some comments. Thank you. > > Cheers, > Andrea > > > -- > Andrea Cristofori > INFN-CNAF > Viale Berti Pichat 6/2 > 40127 Bologna > Italy > Tel. : +39-051-6092920 > Skype: andrea-cnaf > -- Andrea Cristofori INFN-CNAF Viale Berti Pichat 6/2 40127 Bologna Italy Tel. : +39-051-6092920 Skype: andrea-cnaf -- Scanned by iCritical.

1 0

Agenda (and reminder) for today's meeting - 27.11.12 @ 15:00 CET
by Jon Kerr Nilsen 27 Nov '12

27 Nov '12

Hi all, Agenda and a small reminder for today's UR-WG meeting @ 15:00. Agenda: Distribute remaining tasks for UR2.0 draft. Remaining tasks: - Examples (full, job, storage, cloud) - Examples per field - Field summaries (compute, job, memory, storage, cloud) - Proof reading - Decision on cloud block (and would we need use cases?) cheers, Jon

1 0