Efficiency of SACK processing
From Wiki
[edit]
Setup
- See the detailed settings at Testing Setup of kernel 2.6.18-rc4
[edit]
Observation
- We can see very long black-outs in throughput when a lot of packet losses happened in slowstart (especially it was severe with 2.6.18-rc4). We will investigate this problem when first stage of SACK related improvement is done by David.
[edit]
Results
- Bandwidth 400Mbps
| Buffer Size & RTT | Protocols / Kernel Version | 2.6.13 (fastpath) | 2.6.18-rc4 | 2.6.20.3 |
|---|---|---|---|---|
| 0.8*BDP (250ms) | CUBIC | 
| 
| 
|
| HTCP | 
| 
| 
| |
| HSTCP | 
| 
| 
| |
| SACK | 
| 
| 
| |
| 0.8*BDP (380ms) | CUBIC | 
| 
| 
|
| HTCP | 
| 
| 
| |
| HSTCP | 
| 
| 
| |
| SACK | 
| 
| 
| |
| 1.0*BDP (250ms) | CUBIC | 
| 
| 
|
| HTCP | 
| 
| 
| |
| HSTCP | 
| 
| 
| |
| SACK | 
| 
| 
| |
| 1.0*BDP (380ms) | CUBIC | 
| 
| 
|
| HTCP | 
| 
| 
| |
| HSTCP | 
| 
| 
| |
| SACK | 
| 
| 
|
