New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Serverbear's owners pulled the plug out, publishing new service...
End titles for serverbear.com...
Serverscope.io, the owners of this website twitted on the 14th of August that the service is shut down, posting this message:
We shut down ServerBear today, thanks for all the support it was fun!
They released a new server benchmark service in their address, in alpha stage. As they say in their twitter account:
Our priorities for the next release in September:
1. Improved reliability of the benchmark kit
2. Comparison engine for the website
Unfortunately, all the data of the previous benchmarks are gone...
Thanked by 1netomx
Comments
Well, that sucks
Any specific reason o.O
Well damn. I posted several benchmarks there for my own records and comparisons... lol
I suppose, because they presented a new benchmark tool. I tried it but it is buggy...
I cannot understand: why break something that actually works?
Seems a bit shit to just pull the plug like that, it was pretty dead in terms of any updates/development for a long time, seems more to it.
Does anyone have this guys contact?
https://www.lowendtalk.com/profile/serverbear
Domain still has MX records.
Because Stupid..
Well, tbh I never liked it. It was a script very intensive (abusive) which in my opinion didn't represented the real performance of a shared server because it's execution lasted a lot of time.
Also, IIRC the script never was updated, to include new tests or to please some suggestions.
So, in the meanwhile, maybe someone can fork the script and submit the data to another server and start working in a new frontend.
Quite unfortunate. I recall ServerBear was an excellent platform that was frequently used back in the day!
Anyone have any interest in LEB adopting the serverbear idea?
Yep. That would be a nice thing to have.
"These HVH ATOM's seem to outbench these i7 4790k's. Why am I not surprised..."
Francisco
hmm needs some cheap .net domain dam.
I find that a bit ironic coming from you
I'm not sure I follow :P
I'm not sure he knows what 'ironic' means.
https://www.petabyet.com/
I'll be upgrading the backend server in a few days with several new features coming.
They left blog.serverbear.com online.
Nice work there...
Looks horrible on mobile
Hey, just a slight correction. ServerScope is not affiliated with ServerBear.
Here is my initial post about ServerScope: https://lowendtalk.com/discussion/85392/building-a-serverbear-alternative/p1
Btw, before I started working on ServerScope I had reached out to ServerBear owners. My intention was to find a way how I can "resurrect" the website and get it up to date with new design / features. The old site had it's audience even after 2 years of neglect and it would be so much easier to build upon that foundation.
Unfortunately, we couldn't find an option that would work for both of us.
I'm very excited to work on ServerScope these days and brought a couple guys in to help me out with the development. Next release should get us almost to the point of "old" serverbear in terms of functionality and I honestly am putting every spare minute I have into it.
Feel free to reach out to me through email on the website or PM here.
Dammit. The next service I use will have a data download button.
I'd rather run my own DB for my results data if it doesn't.
Also, this is my goto one-liner for testing 4k read/write (4Gb file to override tiny hw-raid caches):
Why would you override RAID cache when benchmarking? When you benchmark the server's performance surely WriteBack cache will increase the I/O performance on like for like machine. Such result would be skewed and would not show the true performance. I don't think it's the way to go.
I love BBU caches on my online.net dedis and zfs ARC(ram) on other dedis.
But when you're trying to fit a larger client database onto a 1-2 GB KVM, I like to know native IO capability.
The correct way, of course, is sizing RAM to match sizeof(' active working set') of the Db and adjusting config.
Anything over 1500 Iops is plenty really for 99% applications.
Err.. Funny may have been a better term to use.
That doesn't really track either.
Did you mean 'appropriate' or 'proper'.
Nope. It seemed funny I guess that @jarland said that considering his background.