blob: 3f49976bd1f8d20b2288bc12beefd896a03c3d73 [file] [log] [blame]
Charles Chan4d5ee192019-09-09 23:08:36 -07001Supported Hardware
2==================
Zack Williamsd63d35b2020-06-23 14:12:46 -07003
4Following is a list of hardware that people from the ONF community have tested
5over time in lab trials.
Charles Chan4d5ee192019-09-09 23:08:36 -07006
7Supported Switches
8------------------
9
101G/10G models (with 40G uplinks)
11^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Zack Williamsd63d35b2020-06-23 14:12:46 -070012
13- OCP Accepted EdgeCore AS5712-54X
14- OCP Accepted EdgeCore AS5812-54X
15- QCT QuantaMesh T3048-LY8
16- Delta AG7648
17- Inventec D6254 (verified by Inventec)
Charles Chan4d5ee192019-09-09 23:08:36 -070018
1925G models (with 100G uplinks)
20^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Zack Williamsd63d35b2020-06-23 14:12:46 -070021
22- QuantaMesh BMS T7032-IX1/IX1B (with 25G breakout cable)
23- Inventec D7054Q28B (verified by Inventec)
Charles Chan4d5ee192019-09-09 23:08:36 -070024
2540G models
26^^^^^^^^^^
Zack Williamsd63d35b2020-06-23 14:12:46 -070027
28- OCP Accepted EdgeCore AS6712-32X
Charles Chan4d5ee192019-09-09 23:08:36 -070029
30100G models
31^^^^^^^^^^^
Zack Williamsd63d35b2020-06-23 14:12:46 -070032
33- OCP Accepted EdgeCore AS7712-32X
34- QCT QuantaMesh BMS T7032-IX1/IX1B
35- OCP Accepted Inventec D7032Q28B (verified by Inventec)
Charles Chan4d5ee192019-09-09 23:08:36 -070036
37Commercially supported models
38^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Zack Williamsd63d35b2020-06-23 14:12:46 -070039
40Models listed here are not supported by ONF community. You should contact
41switch vendor for commercial support.
42
43- OCP Accepted EdgeCore AS5912-54X
Charles Chan4d5ee192019-09-09 23:08:36 -070044
45
46Supported Servers
47-----------------
Zack Williamsd63d35b2020-06-23 14:12:46 -070048ONOS can be run on any x86 servers, with the following notes on CPU and memory
49allocation **per ONOS instance**:
Charles Chan4d5ee192019-09-09 23:08:36 -070050
Zack Williamsd63d35b2020-06-23 14:12:46 -070051- CPU (Recommended: **32 cores**)
52 There are not much requirement on CPU.
Charles Chan4d5ee192019-09-09 23:08:36 -070053
Zack Williamsd63d35b2020-06-23 14:12:46 -070054 In practice, we avoid running other CPU intensive processes/containers/VMs
55 on the same physical machine where ONOS runs.
56
57- Memory (Recommended: **65G RAM**)
58 The memory requirement highly depends on the deployment scale.
59 In general, we suggest that each ONOS instance should get **at least 16G RAM**.
60
61 For production deployment at scale, we will definitely need not only more
62 memory but also some fine tuning on JVM garbage collection mechanism.