... | @@ -296,8 +296,8 @@ The enhancements to current IEEE 802.1AS include |
... | @@ -296,8 +296,8 @@ The enhancements to current IEEE 802.1AS include |
|
[FEC](https://www.ohwr.org/project/white-rabbit/uploads/ff98688542afbe36e9f11877ba4c5f1a/ICALEPCS2011_poster.pdf)),
|
|
[FEC](https://www.ohwr.org/project/white-rabbit/uploads/ff98688542afbe36e9f11877ba4c5f1a/ICALEPCS2011_poster.pdf)),
|
|
~8 "control streams" (defined in VLANs, see
|
|
~8 "control streams" (defined in VLANs, see
|
|
[page 7](https://www.ohwr.org/project/white-rabbit/uploads/1244844e3c3cf850fad84e575f3b3f61/wrCernControlAndTiming.v1.1.pdf))
|
|
[page 7](https://www.ohwr.org/project/white-rabbit/uploads/1244844e3c3cf850fad84e575f3b3f61/wrCernControlAndTiming.v1.1.pdf))
|
|
sent every 1000us, "normal data" size (payload): ~1500 bytes
|
|
sent every 1000us, one-to-many, "normal data" size
|
|
|
|
(payload): ~1500 bytes
|
|
- GSI (not necessarily up-to-date)
|
|
- GSI (not necessarily up-to-date)
|
|
- maximum latency: 100us over ~4 bridge hops @ 1Gbps
|
|
- maximum latency: 100us over ~4 bridge hops @ 1Gbps
|
|
- guaranteed latency over tree-like topology
|
|
- guaranteed latency over tree-like topology
|
... | @@ -307,7 +307,8 @@ The enhancements to current IEEE 802.1AS include |
... | @@ -307,7 +307,8 @@ The enhancements to current IEEE 802.1AS include |
|
200-500 bytes (encoded into 4 frames of size 150-300 bytes
|
|
200-500 bytes (encoded into 4 frames of size 150-300 bytes
|
|
and sent in burst,
|
|
and sent in burst,
|
|
[FEC](https://www.ohwr.org/project/white-rabbit/uploads/ff98688542afbe36e9f11877ba4c5f1a/ICALEPCS2011_poster.pdf)),
|
|
[FEC](https://www.ohwr.org/project/white-rabbit/uploads/ff98688542afbe36e9f11877ba4c5f1a/ICALEPCS2011_poster.pdf)),
|
|
sent every 100us, "normal data" size (payload): ~1500 bytes
|
|
sent every 100us, one-to-many, "normal data" size (payload):
|
|
|
|
~1500 bytes
|
|
- Synchronization: sub-nanosecond in the entire network
|
|
- Synchronization: sub-nanosecond in the entire network
|
|
- Reliability seamless redundancy (synchronization-wise and critical
|
|
- Reliability seamless redundancy (synchronization-wise and critical
|
|
data-wise)
|
|
data-wise)
|
... | @@ -342,25 +343,23 @@ reconfiguration seem to be inline in WR and AVB. (**ToFinish**) |
... | @@ -342,25 +343,23 @@ reconfiguration seem to be inline in WR and AVB. (**ToFinish**) |
|
sufficiently small (i.e. so that maximum 2 frames are lost
|
|
sufficiently small (i.e. so that maximum 2 frames are lost
|
|
during reconfiguration)
|
|
during reconfiguration)
|
|
- two ideas are considered:
|
|
- two ideas are considered:
|
|
- enhanced (HW-supported) Spanning Tree Protocol which allows
|
|
- enhanced Spanning Tree Protocol which allows fast enough
|
|
fast enough reconfiguration
|
|
reconfiguration
|
|
- enhanced Link Aggregation which enables to send frames which
|
|
- enhanced Link Aggregation which enables to send frames which
|
|
constitute the critical data through independent paths
|
|
constitute the critical data through independent paths
|
|
|
|
- any idea needs to support VLANs
|
|
|
|
|
|
### Latency:
|
|
### Latency and scheduling:
|
|
|
|
|
|
- preemption
|
|
|
|
|
|
|
|
<!-- end list -->
|
|
|
|
|
|
|
|
- Time Aware Blocking Shaper:
|
|
|
|
- some problem when cut-through for non-critical traffic -- we
|
|
|
|
don't know the size of the frame
|
|
|
|
|
|
|
|
<!-- end list -->
|
|
|
|
|
|
|
|
|
|
- it need to be taken into account FEC and the fact that "single
|
|
|
|
critical data" translates into many Ethernet Frames
|
|
|
|
- preemption was considered but is currently considered obsolete idea
|
|
|
|
- strict priority scheduling of output queues (Class of Service),
|
|
|
|
resource reservation for critical data and strict control of
|
|
|
|
critical data transmission are considered sufficient for CERN
|
|
|
|
- cut-through forwarding in the switches
|
|
- fault tolerance / isolation
|
|
- fault tolerance / isolation
|
|
- we want to use VLANs to isolate different network and limit a
|
|
- VLANs will be used to isolate different network and limit a
|
|
propagation of destruction by mis-behaving node/switch
|
|
propagation of destruction by mis-behaving node/switch
|
|
- FEC header has ID and sequence number - can prevent some
|
|
- FEC header has ID and sequence number - can prevent some
|
|
duplication issues
|
|
duplication issues
|
... | @@ -374,3 +373,9 @@ reconfiguration seem to be inline in WR and AVB. (**ToFinish**) |
... | @@ -374,3 +373,9 @@ reconfiguration seem to be inline in WR and AVB. (**ToFinish**) |
|
then one source of time, it is possible to make
|
|
then one source of time, it is possible to make
|
|
voting/comparision)
|
|
voting/comparision)
|
|
|
|
|
|
|
|
<!-- end list -->
|
|
|
|
|
|
|
|
- Time Aware Blocking Shaper:
|
|
|
|
- some problem when cut-through for non-critical traffic -- we
|
|
|
|
don't know the size of the frame
|
|
|
|
|