Thor vs. orin: sysbench memory run

Robotics & Edge Computing Jetson Systems Jetson Thor
1

Thor: 128GB 256-bit LPDDR5X 273GB/s, but test result: 3230.93 MiB/sec

Orin: 64GB 256-bit LPDDR5 204.8GB/s, but test result: 3963.17 MiB/sec

then why thor slower than orin?

in Thor with jetpack 7.0:

nvidia@thor:~$ nvpmodel -q
NV Power Mode: MAXN
0

nvidia@thor:~$ sudo jetson_clocks

nvidia@thor:~$ sysbench memory run
sysbench 1.0.20 (using system LuaJIT 2.1.0-beta3)

Running the test with following options:
Number of threads: 1
Initializing random number generator from current time

Running memory speed test with the following options:
block size: 1KiB
total size: 102400MiB
operation: write
scope: global

Initializing worker threads…

Threads started!

Total operations: 33089226 (3308475.08 per second)

32313.70 MiB transferred (3230.93 MiB/sec)

General statistics:
total time: 10.0000s
total number of events: 33089226

Latency (ms):
min: 0.00
avg: 0.00
max: 0.04
95th percentile: 0.00
sum: 3598.69

Threads fairness:
events (avg/stddev): 33089226.0000/0.00
execution time (avg/stddev): 3.5987/0.00

and in orin with jetpack 6.2:

nvidia@orin:~$ nvpmodel -q
NV Power Mode: MAXN
0

nvidia@orin:~$ sudo jetson_clocks

nvidia@orin:~$ sysbench memory run
sysbench 1.0.20 (using system LuaJIT 2.1.0-beta3)

Running the test with following options:
Number of threads: 1
Initializing random number generator from current time

Running memory speed test with the following options:
block size: 1KiB
total size: 102400MiB
operation: write
scope: global

Initializing worker threads…

Threads started!

Total operations: 40588655 (4058289.95 per second)

39637.36 MiB transferred (3963.17 MiB/sec)

General statistics:
total time: 10.0001s
total number of events: 40588655

Latency (ms):
min: 0.00
avg: 0.00
max: 0.01
95th percentile: 0.00
sum: 4323.77

Threads fairness:
events (avg/stddev): 40588655.0000/0.00
execution time (avg/stddev): 4.3238/0.00

3

Hi,
Thanks for reporting this. We will check it.