blob: cc1bf8dd1c17b18bed1e31af17cdbfa3970a5ff4 (
plain)
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
|
CSIT Release Notes
==================
Changes in CSIT |release|
-------------------------
#. Added Testpmd tests
- new NICs - Intel x520
Multi-Thread and Multi-Core Measurements
----------------------------------------
**HyperThreading** - CSIT |release| performance tests are executed with SUT
servers' Intel XEON CPUs configured in HyperThreading Disabled mode (BIOS
settings). This is the simplest configuration used to establish baseline
single-thread single-core SW packet processing and forwarding performance.
Subsequent releases of CSIT will add performance tests with Intel
HyperThreading Enabled (requires BIOS settings change and hard reboot).
**Multi-core Test** - CSIT |release| multi-core tests are executed in the
following Testpmd thread and core configurations:
#. 1t1c - 1 Testpmd pmd thread on 1 CPU physical core.
#. 2t2c - 2 Testpmd pmd threads on 2 CPU physical cores.
#. 4t4c - 4 Testpmd pmd threads on 4 CPU physical cores.
Note that in many tests running Testpmd reaches tested NIC I/O bandwidth
or packets-per-second limit.
Packet Throughput Measurements
------------------------------
Following values are measured and reported for packet throughput tests:
- NDR binary search per RFC2544:
- Packet rate: "RATE: <aggregate packet rate in packets-per-second> pps
(2x <per direction packets-per-second>)"
- Aggregate bandwidth: "BANDWIDTH: <aggregate bandwidth in Gigabits per
second> Gbps (untagged)"
- PDR binary search per RFC2544:
- Packet rate: "RATE: <aggregate packet rate in packets-per-second> pps (2x
<per direction packets-per-second>)"
- Aggregate bandwidth: "BANDWIDTH: <aggregate bandwidth in Gigabits per
second> Gbps (untagged)"
- Packet loss tolerance: "LOSS_ACCEPTANCE <accepted percentage of packets
lost at PDR rate>""
- NDR and PDR are measured for the following L2 frame sizes:
- IPv4: 64B, 1518B, 9000B.
Packet Latency Measurements
---------------------------
TRex Traffic Generator (TG) is used for measuring latency of Testpmd DUTs.
Reported latency values are measured using following methodology:
- Latency tests are performed at 10%, 50% of discovered NDR rate (non drop rate)
for each NDR throughput test and packet size (except IMIX).
- TG sends dedicated latency streams, one per direction, each at the rate of
10kpps at the prescribed packet size; these are sent in addition to the main
load streams.
- TG reports min/avg/max latency values per stream direction, hence two sets
of latency values are reported per test case; future release of TRex is
expected to report latency percentiles.
- Reported latency values are aggregate across two SUTs due to three node
topology used for all performance tests; for per SUT latency, reported value
should be divided by two.
- 1usec is the measurement accuracy advertised by TRex TG for the setup used in
FD.io labs used by CSIT project.
- TRex setup introduces an always-on error of about 2*2usec per latency flow -
additonal Tx/Rx interface latency induced by TRex SW writing and reading
packet timestamps on CPU cores without HW acceleration on NICs closer to the
interface line.
Report Addendum Tests - Additional NICs
---------------------------------------
Adding test cases with more NIC types. Once the results become available, they
will be published as an addendum to the current version of CSIT |release|
report.
|
