You love benchmarks. We love benchmarks. The LowEnd community can’t get enough benchmarks!
One of the big reasons people constantly run and post them is that it helps others determine if a service is going to work for their needs. If you are looking for a home for your game server and you live in the middle of Washington, Seattle seems logical…but not if the provider’s network is back-hauled to Los Angeles.
YABS has emerged over the years as the community favorite. In fact, I took some heat a while back for saying it couldn’t benchmark things I was interested in the most, but that was from people who didn’t read the full article (my point is that no benchmark can tell you about reliability, how a provider handles tickets, provider security, etc.). We’ve covered other benchmarkers, too.
But YABS is practically a common noun like google. “Please post a YABS”. That benchmark will give you basic system info (everything from if AES-NI is available to kernel version), disk speeds, network performance, and CPU tests.
network-speed
So when LowEndTalk member @sh97 contacted me to run a piece on his new benchmarking script – network-speed – my first question was “what does it do that YABS doesn’t”?
He said “The biggest difference is that YABS is an all-in-one bench script – CPU, Network (iperf) and Storage speed, while network-speed focuses only on Network (speedtest-ookla)”.
Now @sh97 wrote the network info part of YABS, so he’s a bit of a specialist in this field. network-speed is likewise a specialist, for a very good reason:
The biggest problem I had, as a person from Asia/India is that I can never measure the network performance using YABS since there are no nearby servers and just a single Asian location which is nearer to EU than Asia. And since iperf can handle just a single connection at a time, during peak hours it’s impossible to run a full network test. Now, I just need to run an India bench to check the performance to all major networks of India (as an example).
Nice! Via Ookla (speedtest.net), you can check your network in North and South America, Europe, Asia, Africa, GCC Middle East, India, China, and even Iran, with more coming soon.
As I said, you love benchmarks, so check this one out at network-speed.xyz.
Thanks for the tip @sh97! I’m sure he’d love some feedback and suggestions for improvements, or just a friendly thumbs-up if you find it useful.
Related Posts:
I also enjoy using LEBs! I have an empire of little guys for VPNs, larger guys for apps, and big guys for databases and storage. When I'm not in front a screen I'm into German Shepherd dogs, quality knives, target shooting, theology, tabletop roleplaying games, and forest hiking.
I enjoy writing technical articles here on LowEndBox to help people get more out of their systems. You can find me on LowEndTalk @raindog308
I am a person that likes everything up front. So, to all that try this, your results are uploaded to some site with a .ch extension. Example:
https://imgur.com/o8i2kxy
I do not see my IP listed, but a lot of information about your server is stored on this site https://cdn1.frocdn.ch/. I used my local dev server and here is just part of what is stored in a .txt file:
—————————————————————————
Basic System Info
—————————————————————————
CPU Model : AMD A8-6500B APU with Radeon(tm) HD Graphics
CPU Cores : 4 @ 2165.365 MHz
CPU Cache : 2048 KB
AES-NI : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
Total Disk : 458.0 GB (42.9 GB Used)
Total RAM : 14.6 GB (1.9 GB Used)
Total Swap : 7.5 GB (2.1 MB Used)
System uptime : 8 days, 23 hour 16 min
Load average : 0.00, 0.00, 0.00
OS : AlmaLinux release 8.8 (Sapphire Caracal)
Arch : x86_64 (64 Bit)
Kernel : 4.18.0-477.10.1.el8_8.x86_64
Virtualization : NONE
======================
Not saying this is all bad, but if you are like me, you are very security conscious.
This script is great…???
I ran the “curl -sL network-speed.xyz | bash”
The dedicated server I was using this on suddenly stopped the network test and a message in ssh said “suspended”
I couldn’t ping the server anymore and any vpn I used could not access ssh anymore.
Out of curiosity, what does the server owners look at, which makes them think that I was doing something wrong to suspend me?
We’ve noticed some providers (especially some in EU) mistake running this script as an outbound DDOS attack, so all access would be cut off. Won’t entirely blame them, since the port is being maxed out for 10-15 minutes as the tests are run.
Tbh, there is not much info there such that a server would be at risk. As you pointed out, IP is not being stored or shown anywhere.
Regarding frocdn, it is operated by Advin Servers, a provider who is present on LowEndTalk. Again, I think it’s not an issue. Anyways, we recently switched to an in-house storage system, so now files are stored on a VPS instead of a 3rd party host.
In the coming months, we will add the option to make result sharing optional.
Great article. Thanks