| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
|
All, our editors have done a mammoth job
addressing all of the comments submitted against Draft 0.2 in response to the
mammoth job many of you have done in reviewing that draft. I thank you all
for your efforts to date. While the editors prepare Draft 0.3 many of us
are actively participating in various Ad Hoc groups to continue to resolve open
issues. There is an additional area where contributions are welcome.
As we examine the final comment resolution data-base it becomes apparent that
some of the comment resolutions request further input. Let me provide one
example:
Jason Fan's Technically binding comment #90
states:
Concern:
A group needs to be tasked to generate a clause for
ringlet selection. This section should, among other things, address:
1. The automatic selection of the appropriate ringlet for unicast traffic (both in normal and ring protection scenarios); 2. The automatic selection of both ringlets for multicast traffic in steering protection scenarios. Also how the packet is replicated to both ringlets and how the TTL is set appropriately; 3. The explicit selection by the client of which ringlet the traffic should be transmitted on in normal or protection scenarios; 4. The algorithm(s) used to map each packet to the appropriate ringlet or ringlets on which it should be transmitted; 5. The handling of unprotected traffic, e.g such traffic may be dropped during protection scenarios. with his suggested
remedy:
No clause is addressing ringlet selection, shown as
a necessary component on the add data path in the MAC in Section
5.8
The Group resolution is as
follows:
This comment appears to relate to ringlet selection
during (a) normal mode, and (b) protection mode,
for various traffic scenarios. There is also some need outlined to discuss packet replication and TTL setting. This appears to be a deficiency in the current draft and should be fixed. Contributions and proposals on the subject are invited from the WG; if these contributions indicate that the deficiency exists and that a remedy should be found, then the appropriate clauses will be modified. However, the "ringlet selection" clause was dropped by the WG during initial draft creation as the information that would go within it was felt to be better placed in other clauses (the MAC Datapath or Protection clauses). Hence it is recommended that in the above scenario, the MAC Datapath or the Protection clauses each include a subclause titled "Ringlet selection" that specifies how ringlets are selected during normal and protection modes, when the information is available. This remedy invites contributions from the
WG. Please review the final comment resolutions. Understand that
where no answer is provided, further inputs will be required to complete the
draft. Creation of these inputs can begin while our editors are preparing
Draft 0.3.
Thank you for all your efforts to date.
Please continue to focus on the many areas of the draft that still require
improvement. The final resolution of Draft 0.2 comments will provide
additional guidance on where some of those areas are.
Best regards,
Robert D. Love
President, Resilient Packet Ring Alliance President, LAN Connect Consultants 7105 Leveret Circle Raleigh, NC 27615 Phone: 919 848-6773 Mobile: 919 810-7816 email: rdlove@xxxxxxxx Fax: 208 978-1187 |