VSC7440 datasheet(244/444 Pages) MICROSEMI

244 / 444 page

Functional Descriptions

VMDS-10492 VSC7440 Datasheet Revision 4.2

227

•

Data rate shapers are connected in a linked list, which is traversed every 10 µs. The shaper rate

granularity becomes 100 kbps.

•

Frame rate shapers are connected in a linked list, which is traversed every 1 ms. The shaper rate

granularity becomes 1 kilo frames per second.

Each scheduler element includes two leaky buckets to support DLB shaping. The committed information

rate (CIR) bucket is open when the filling is below the configured limit. The excess information rate (EIR)

shaper is always closed, unless the congestion control reports that a threshold is reached. Which

congestion thresholds to react on are configurable for each shaper and should depend on the traffic

groups the particular element is a part of. The effect of the DLB shaper is that the rate out of the shaper

is CIR or CIR plus EIR, depending of the accumulation of data in the queue system.

The following table lists the registers associated with shaping.

3.24.7.4 Bandwidth Distribution

Each scheduler element is selecting one of its inputs as the next source of transmission if a scheduling

process passes through it. A number of different options are available for how this arbitration takes place.

A configured value for the scheduler element tells how many of the lower indexed inputs should be

arbitrated with the DWRR method. Inputs with higher indexes are preferred and arbitrated strict.

Figure 68 •

Scheduler

The lower indexes are arbitrated in the effort to achieve a bandwidth ratio between them. In this regard,

the bandwidth can be measured in frames, data bits, or line bits (including IFG). Each input is configured

with a cost setting, and the algorithm strives to give n times more bandwidth to input A than B, if the cost

configuration for A is n times lower than B. The following are two cost configuration examples.

•

Input 4 has cost 7, and input 6 has cost 11. If the total bandwidth shaped out of the scheduler

element is 180 Mbps, input 4 will get 110 Mbps of it.

•

All inputs have cost 1, and the element is configured for frames per second arbitration. Algorithm will

by this strive to send an equal number of frames from each input.

Table 153 •

Shaper Registers Overview

Description

HSCH::SE_CFG

Selects filling mode

HSCH::CIR_CFG

Configures leak rate and burst capacity for CIR bucket

HSCH::EIR_CFG

Configures leak rate and burst capacity for EIR bucket

HSCH::SE_DLB_SENSE

Selects thresholds to control EIR with.

HSCH::HSCH_LEAK_LIST

Configures leak list periods and starting index

HSCH::SE_CONNECT

Links to next in leaking lists

HSCH::CIR_STATE

Returns current CIR bucket filling

HSCH::EIR_STATE

Returns current EIR bucket filling

HSCH::SE_STATE

Returns whether or not the shaper is open

VSC7440 Datasheet(PDF) 244 Page - Microsemi Corporation

VSC7440 Datasheet(HTML) 244 Page - Microsemi Corporation

Html Pages

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 ...More

Datasheet Download

Link URL