tree configuration in bridged ieee1394 bus network
Post on 06-Feb-2016
41 Views
Preview:
DESCRIPTION
TRANSCRIPT
Tree Configuration in Bridged Tree Configuration in Bridged IEEE1394 Bus NetworkIEEE1394 Bus Network
PHILIPS Research
Subrata Banerjee
PHILIPS Research Briarcliff
P1394.1 WG MeetingMarch 19-20, 1998
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
Problem StatementProblem Statement
Bridges of different capabilities Bandwidth Iso Delay
Bridges may introduce unacceptable loops
Path between any two bridges “Minimum bottleneck” route
Typically bridge capacity lower than bus capacity
N1 C1 C2 N2
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
Graph Theory ResultGraph Theory Result
Maximum Spanning Tree guarantees
Minimum Bottleneck Route
Between every pair of nodes
Distributed maximum spanning tree algorithm
C2
C1
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
How to Choose Between Two Bridges?How to Choose Between Two Bridges?
Bridge Capability Parameters Bridge Bandwidth
Bridge Iso_Delay
Bridge Vendor ID = max. of 2 portal vendor IDs
Bridge Node ID = max. of 2 portal vendor IDs
Proposed Organization
Bridge BW (13) Resv (3) Iso_delay (8) Resv (8)
Chip ID (40)
Vendor ID (24) (scrambled)
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (1)(1)
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
5827
48
57
3836
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
5827
48
57
3836
BP62
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (2)(2)
BP Broadcasts
BP31
BP40 BP52
BP65
BP27
BP36BP38
BP48
BP21
BP57 BP55
BP58
BP34
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
5827
48
57
3836
BP Count=2
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (3)(3)
Fragment RootsBP Count=3
BP Count=3
BP Count=2BP Count=4
62, L1
58, L1
57, L1
48, L1
65, L1
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
5827
48
57
3836
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (4)(4)
Find Best Neighbor
62, L1
58, L1
57, L1
48, L1
65, L1
Submit
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
5827
48
57
3836
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (5)(5)
Join 1
62, L1
58, L1
57, L1
48, L1
65, L1
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
5827
48
57
3836
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (6)(6)
Join 1
58, L1
57, L1
48, L1
65, L2
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
5827
48
57
3836
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (7)(7)
Find new best neighbor
58, L1
57, L1
48, L1
65, L2
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
5827
48
57
3836
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (8)(8)
Join 2
58, L1
57, L1
48, L1
65, L2
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
5827
48
57
3836
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (9)(9)
Join 2
58, L2
48, L1
65, L2
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
5827
48
57
3836
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (10)(10)
Join 3
48, L1
65, L2 58, L2
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
5827
48
57
3836
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (11)(11)
Join 3
58, L265, L3
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
5827
48
57
36
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (12)(12)
Join 3
58, L265, L3
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
5827
48
57
36
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (13)(13)
Join 4
58, L2
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
31
38
48
40
27
21
3634
58
55
65
58
48
57
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (14)(14)
Join 4
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
62
65 52
57
48
40
58
55
65
58
48
57
An Example of Tree Conf. Algorithm An Example of Tree Conf. Algorithm (15)(15)
Join 4
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
Rules of the GameRules of the Game
Fragment A can submit to Fragment B iffLevelA LevelB
Bridge Capability values are unique.
No loops possible
Lx Ly
Lz
Lx > Lz > Ly > Lx
Lx Lx
LxC3
C2
C1
C1 > C2 > C3 > C1
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
Example of Selected CommandsExample of Selected Commands
“I am a BP” data = Unique Bridge Capabilities (UBC)
“Report UBC” data = best UBC from all children
“Connect” data = Fragment Level
“Update” data = Fragment UID, Fragment Level
PHILIPS Subrata Banerjeesub@philabs.research.philips.com
Once the Tree Topology is Identified ...Once the Tree Topology is Identified ...
Assign Bus Ids
Routing Bounds
Portal_Control.rte fields
Select Net Cycle Timer
Assign Portal_Control.clk fields
Details?
top related