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.
SwitchVM - Benchmarks?
Hello,
SwitchVM will be launching extremely soon, and I'd like 2-3 people to do some benchmarks for me, as I know you like doing that!
If you would like to take part, just reply to this post and I'll PM you with details.
Thanks guys!
Dominic
SwitchVM
This discussion has been closed.
Comments
Benchmark on an empty node? Big deal.
I didn't say somebody had to do them. I was asking if somebody could
What makes you think this will work better than GuernseyHosting?
But what would you get by benchmarking an empty node?
Astounding!
Because Guernsey Hosting didn't have enough money to start up, and therefore was making a loss. Even if I am going to make a loss to start with (which all hosts do), I will have the money to carry on renewing.
We aren't reselling VPS's (when GSYHosting did, it was painful).
Results. I just want to see other people think it performs for them, and how it does perform
Empty node Benchmark != Real usage, once you atleast feel it upto 50%, that
benchmark will only be able to prove something.
I might be wrong.
benchmark will only be able to prove something.
I might be wrong.
Ah OK, fair enough. I'll wait until i'm 50-60% full and then ask a few to benchmark then
I agree.
We were getting over 240MB/s Disk I/O on an empty node, but only around 180MB/s now that our node is 75% full.
benchmark will only be able to prove something.
I might be wrong.
You are not wrong. The point of the initial benchmarks* on empty node is to check if everything is actually okey with the hardware (for example RAID10 node, having 50MB/s read/write speeds on dd test) and eventually to compare once it is empty and once it is full. That should be done by 1 person only, the owner/staff member.
I did a KVM beta testing a month ago - the purpose was not benchmarks, but ot see how the node performs under load, which is something that most hosts should do, when stepping into new waters (as it was KVM for me).
Thanks for the input guys. I'll get somebody to test when this node is about 50-60% full.
Thanks for the laugh dom
If you are going to be feeling up the node, please take it to the server porn thread
My bad @mitGib fill. Auto correct strikes again.
Me me me...
Tim is sharp
Sounds funny
I'm still waiting for this topic:
"SwitchVM - Stock??"
You should be benchmarking yourself before launch imo. That's the only way to tune performance.
If your using SecuredServers... then just go get a gigabit port for your VPS node. It's only $5.
I got a free upgrade to a E3-1270v2 from a 1230v1 because I told Gordon his network seemed amazing.
Wow, just want to say that your network is also seems amazing
and even more than amazing ^ ^
I will be getting gbit. Don't worry
What you node configuration?
This has been said..so..many..times
First node is an X3440 with 2xHDD's in SWRAID 1. Second node will be an E3-1230v2 with 4xHDD's in RAID10.
Thanks!
which dc?
PhoenixNAP. I will get more US locations, eventually.
@dominicl How much RAM? Also, please get RAID10 on Both nodes
This dc have nice connection from South East Asia.
Also nice connection to New Zealand
160ms at 2AM, 260ms at "peak" times.
care to share?