benchmark: update num to n in dgram offset-length

[brunocroh]

Replace the num config with n to be compatible with the n CLI argument.

According to nodejs/performance#186 this benchmark was taking 40 seconds for a single run.

Since this benchmark doesn’t take too long to run (about 40s), I didn’t change the config.

I ran the benchmarks for comparison, but I’m not sure if using len as [1, 512, 1024] would provide better coverage than [1, 64, 256, 1024]. What do you guys think?

If that makes sense to you, let me know, and I can update it to [1, 512, 1024].
This change reduces the execution time from 40s to 30s.

Here’s the output for both configs:

len: [1, 64, 256, 1024]

dgram/offset-length.js dur=5 type="send" n=100 len=1: 0.00029774464483540395
dgram/offset-length.js dur=5 type="recv" n=100 len=1: 0.00008736534416572343
dgram/offset-length.js dur=5 type="send" n=100 len=64: 0.01820060746926647
dgram/offset-length.js dur=5 type="recv" n=100 len=64: 0.005719775647413869
dgram/offset-length.js dur=5 type="send" n=100 len=256: 0.07849402316926338
dgram/offset-length.js dur=5 type="recv" n=100 len=256: 0.02109259972351112
dgram/offset-length.js dur=5 type="send" n=100 len=1024: 0.3289801256746718
dgram/offset-length.js dur=5 type="recv" n=100 len=1024: 0.09215122503886834

real    0m40.983s
user    0m21.080s
sys     0m17.143s

len: [1,512,1024]

dgram/offset-length.js dur=5 type="send" n=100 len=1: 0.0003172381715335142
dgram/offset-length.js dur=5 type="recv" n=100 len=1: 0.00010832342823154378
dgram/offset-length.js dur=5 type="send" n=100 len=512: 0.1707906782931467
dgram/offset-length.js dur=5 type="recv" n=100 len=512: 0.05277125834012046
dgram/offset-length.js dur=5 type="send" n=100 len=1024: 0.3028087268917686
dgram/offset-length.js dur=5 type="recv" n=100 len=1024: 0.08280481519975962

real    0m30.869s
user    0m15.585s
sys     0m13.336s

Benchmark Machine Specifications:

---------------------------------
Uptime     : 3 days, 15 hours, 36 minutes
Processor  : Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz
CPU cores  : 4 @ 2999.998 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM        : 11.4 GiB
Swap       : 0.0 KiB
Disk       : 500.0 GiB
Distro     : CentOS Stream 9
Kernel     : 5.14.0-295.el9.x86_64
VM Type    : KVM

[nodejs-github-bot]

Review requested:

• @nodejs/performance

[RafaelGSS]

If that makes sense to you, let me know, and I can update it to [1, 512, 1024].

+1

[brunocroh]

If that makes sense to you, let me know, and I can update it to [1, 512, 1024].

+1

done

[RafaelGSS]

@brunocroh can you make sure to update the spreadsheet with the PRs you did? Also, let's focus just on benchmarks that take more than 10 min (on compare.js) to conclude

[brunocroh]

@brunocroh can you make sure to update the spreadsheet with the PRs you did? Also, let's focus just on benchmarks that take more than 10 min (on compare.js) to conclude

Yes, sure. Since I don’t have permission to edit the table, I just added comments with the PR URL.

Regarding the duration, I’m trying to focus on the ones that take longer to execute. At the same time, I’m also updating the ones that don’t support the N flag because of the requirements in nodejs/performance#187.