Hi everyone I've reviewed section 4, 5, 6 in the document. Have attached the document (based on v14) with tracking changes (next time can we do LaTeX please :-)) Some comment with the most controversial stuff: Some write startTime and other start_time (and similar with other things). I've corrected most to "start time", however I think the most important think is consistency, so if someone feels it should be one over the other, go ahead, but please just keep it consistent. Remember this is the CS protocol, not the NSI protocol. Section 4.2: I've changed the title from provision to activation, but more importantly changed some text about what to set start time to if immediate activation is required, so please review that if you have a bone in that. Section 5: In general, very short sections, which makes it seem a bit different. Some of them can probably be joined. 5.4: I have remark about correlationId, I don't think it is needed at coordinator level, only at the MTL. 5.5: I am not sure what the purpose of the list of timeout causes is. I'd merge 5.7 and 5.8 Finally, do we call it "a reservation" or "a connection". We are not consistent with this at all. My take is that is "a connection", as a reservation is only a subset of what it is. There are also quite a few changes and some comments for section 5. Chin, please review. /Henrik
Hi all, I did a once over the document and made some minor edits. I tried to address Henrik's comments in Sec 5, and also accepted most of his changes. There are still a few outstanding changes that need to be done (i.e. merging tables in Sec 3.) Please feel free to accept/reject the changes. Thanks! - Chin On 5/30/13 5:12 AM, Henrik Thostrup Jensen wrote:
Hi everyone
I've reviewed section 4, 5, 6 in the document. Have attached the document (based on v14) with tracking changes (next time can we do LaTeX please :-))
Some comment with the most controversial stuff:
Some write startTime and other start_time (and similar with other things). I've corrected most to "start time", however I think the most important think is consistency, so if someone feels it should be one over the other, go ahead, but please just keep it consistent.
Remember this is the CS protocol, not the NSI protocol.
Section 4.2: I've changed the title from provision to activation, but more importantly changed some text about what to set start time to if immediate activation is required, so please review that if you have a bone in that.
Section 5: In general, very short sections, which makes it seem a bit different. Some of them can probably be joined.
5.4: I have remark about correlationId, I don't think it is needed at coordinator level, only at the MTL.
5.5: I am not sure what the purpose of the list of timeout causes is.
I'd merge 5.7 and 5.8
Finally, do we call it "a reservation" or "a connection". We are not consistent with this at all. My take is that is "a connection", as a reservation is only a subset of what it is.
There are also quite a few changes and some comments for section 5. Chin, please review.
/Henrik
_______________________________________________ nsi-wg mailing list nsi-wg@ogf.org https://www.ogf.org/mailman/listinfo/nsi-wg
-- Chin Guok NOC: (510) 486-7600 ESnet Network Engineering Group (AS293) (800) 333-7638 Lawrence Berkeley National Laboratory
Chin, Henrik, Thanks for these edits, I will make this document the basis for version 15 and upload this to redmine.ogf.org shortly. Regards, Guy From: nsi-wg-bounces@ogf.org [mailto:nsi-wg-bounces@ogf.org] On Behalf Of Chin Guok Sent: 03 June 2013 11:11 To: nsi-wg@ogf.org Subject: Re: [Nsi-wg] More document reviewing Hi all, I did a once over the document and made some minor edits. I tried to address Henrik's comments in Sec 5, and also accepted most of his changes. There are still a few outstanding changes that need to be done (i.e. merging tables in Sec 3.) Please feel free to accept/reject the changes. Thanks! - Chin On 5/30/13 5:12 AM, Henrik Thostrup Jensen wrote: Hi everyone I've reviewed section 4, 5, 6 in the document. Have attached the document (based on v14) with tracking changes (next time can we do LaTeX please :-)) Some comment with the most controversial stuff: Some write startTime and other start_time (and similar with other things). I've corrected most to "start time", however I think the most important think is consistency, so if someone feels it should be one over the other, go ahead, but please just keep it consistent. Remember this is the CS protocol, not the NSI protocol. Section 4.2: I've changed the title from provision to activation, but more importantly changed some text about what to set start time to if immediate activation is required, so please review that if you have a bone in that. Section 5: In general, very short sections, which makes it seem a bit different. Some of them can probably be joined. 5.4: I have remark about correlationId, I don't think it is needed at coordinator level, only at the MTL. 5.5: I am not sure what the purpose of the list of timeout causes is. I'd merge 5.7 and 5.8 Finally, do we call it "a reservation" or "a connection". We are not consistent with this at all. My take is that is "a connection", as a reservation is only a subset of what it is. There are also quite a few changes and some comments for section 5. Chin, please review. /Henrik _______________________________________________ nsi-wg mailing list nsi-wg@ogf.org<mailto:nsi-wg@ogf.org> https://www.ogf.org/mailman/listinfo/nsi-wg -- Chin Guok NOC: (510) 486-7600 ESnet Network Engineering Group (AS293) (800) 333-7638 Lawrence Berkeley National Laboratory
participants (3)
-
Chin Guok -
Guy Roberts -
Henrik Thostrup Jensen