All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Which one is better to retain
Merlincool
Member
Hey folks I have two vpses and I need to stop renewal of one of them, which one you would recommend to keep. Also it would be better if you can offer explanation which one is better. I am mainly using this two for torrenting and Occasional streaming using jellyfin, so which is better over other and by what did you came to a conclusion.
**VPS 1 Benchmark OVZ **
`# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Yet-Another-Bench-Script
v2020-12-29
https://github.com/masonr/yet-another-bench-script
## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##
Fri 19 Mar 06:10:45 UTC 2021
Basic System Information:
Processor : Intel(R) Xeon(R) CPU E3-1240 v6 @ 3.70GHz
CPU cores : 2 @ 3696.261 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM : 768.0 MiB
Swap : 512.0 MiB
Disk : 400.0 GiB
fio Disk Speed Tests (Mixed R/W 50/50): ---------------------------------
| Block Size | 4k (IOPS) | 64k (IOPS) |
|---|---|---|
| Read | 93.58 MB/s (23.3k) | 330.79 MB/s (5.1k) |
| Write | 93.83 MB/s (23.4k) | 332.54 MB/s (5.1k) |
| Total | 187.41 MB/s (46.8k) | 663.33 MB/s (10.3k) |
| Block Size | 512k (IOPS) | 1m (IOPS) |
| ------ | --- ---- | ---- ---- |
| Read | 360.86 MB/s (704) | 459.07 MB/s (448) |
| Write | 380.04 MB/s (742) | 489.64 MB/s (478) |
| Total | 740.91 MB/s (1.4k) | 948.72 MB/s (926) |
iperf3 Network Speed Tests (IPv4):
Provider | Location (Link) | Send Speed | Recv Speed
| | |
Clouvider | London, UK (10G) | 929 Mbits/sec | 936 Mbits/sec Online.net | Paris, FR (10G) | 891 Mbits/sec | 930 Mbits/sec WorldStream | The Netherlands (10G) | 913 Mbits/sec | busy Biznet | Jakarta, Indonesia (1G) | 467 Mbits/sec | busy Clouvider | NYC, NY, US (10G) | 607 Mbits/sec | 741 Mbits/sec Velocity Online | Tallahassee, FL, US (10G) | 678 Mbits/sec | 212 Mbits/sec Clouvider | Los Angeles, CA, US (10G) | 596 Mbits/sec | 422 Mbits/sec Iveloz Telecom | Sao Paulo, BR (2G) | 456 Mbits/sec | 696 Mbits/sec
Running GB5 benchmark test... cue elevator music`
VPS 2 BENCHMARK KVM Vps
`# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Yet-Another-Bench-Script
v2020-12-29
https://github.com/masonr/yet-another-bench-script
## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##
Fri 19 Mar 06:15:45 UTC 2021
Basic System Information:
Processor : Intel(R) Xeon(R) CPU E3-1230 V2 @ 3.30GHz
CPU cores : 1 @ 3300.022 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 987.2 MiB
Swap : 1023.0 MiB
Disk : 503.0 GiB
fio Disk Speed Tests (Mixed R/W 50/50):
| Block Size | 4k (IOPS) | 64k (IOPS) |
|---|---|---|
| Read | 3.40 MB/s (852) | 41.80 MB/s (653) |
| Write | 3.44 MB/s (860) | 42.05 MB/s (657) |
| Total | 6.84 MB/s (1.7k) | 83.85 MB/s (1.3k) |
| Block Size | 512k (IOPS) | 1m (IOPS) |
| ------ | --- ---- | ---- ---- |
| Read | 119.11 MB/s (232) | 125.28 MB/s (122) |
| Write | 125.44 MB/s (245) | 133.63 MB/s (130) |
| Total | 244.56 MB/s (477) | 258.92 MB/s (252) |
iperf3 Network Speed Tests (IPv4):
Provider | Location (Link) | Send Speed | Recv Speed
| | |
Clouvider | London, UK (10G) | 942 Mbits/sec | 931 Mbits/sec
Online.net | Paris, FR (10G) | 934 Mbits/sec | 933 Mbits/sec
WorldStream | The Netherlands (10G) | busy | 934 Mbits/sec
Biznet | Jakarta, Indonesia (1G) | busy | 363 Mbits/sec
Clouvider | NYC, NY, US (10G) | 893 Mbits/sec | 803 Mbits/sec
Velocity Online | Tallahassee, FL, US (10G) | 851 Mbits/sec | 262 Mbits/sec
Clouvider | Los Angeles, CA, US (10G) | 811 Mbits/sec | 739 Mbits/sec
Iveloz Telecom | Sao Paulo, BR (2G) | 786 Mbits/sec | 505 Mbits/sec
iperf3 Network Speed Tests (IPv6):
Provider | Location (Link) | Send Speed | Recv Speed
| | |
Clouvider | London, UK (10G) | 920 Mbits/sec | 920 Mbits/sec
Online.net | Paris, FR (10G) | 919 Mbits/sec | 915 Mbits/sec
WorldStream | The Netherlands (10G) | 915 Mbits/sec | 920 Mbits/sec
Clouvider | NYC, NY, US (10G) | 874 Mbits/sec | 864 Mbits/sec
Clouvider | Los Angeles, CA, US (10G) | 814 Mbits/sec | 787 Mbits/sec`
Comments
first one?
The first one: 2 vcpu + faster disk speed
First one is OVZ vps and later is KVM. I think KVM is much better over OVZ
The first one seems better
I'd take the one with more RAM
Monitoring your cpu and ram usage could help. What about TOS. Are there anything "bad"? Price?
If both are working fine keep cheaper.
Which one works for you?
Personal opinion, I personally hate OVZ with a passion, so I'd say get rid of the OVZ one.. Business opinion: keep whichever meets your needs the best.
Both are of equal prices
main part is torrenting, I haven't seen any difference between two, just that both are incapable of making 1:1 ratio
Both works but as any of you have some good and better experience in knowing what all features I must go for torrenting.
How to know or identify correct right and read speed, there are various block size described, so what does that means
This is the kind of question you should just google and find out yourself. To put it simply, you read and write in blocks of data, the size is just a parameter used in the fio tests. Different use cases may benefit from different block sizes, but if you are asking this question it is probably not important to you.
Google block size and learn more if you want
Honestly, if you can install all the apps you want on the ovz, why not just stick with it? You're fine as long as it isn't ovz6 or older.
That may depend a lot on where you are tracking those torrents. If it is private trackers where there are lots of people with high speed lines it can be difficult because by the time you have the content so does most of the rest of the swarm so there are only a few still needing bits and lots of places they can download them from.
Depending on how your provider manages your bandwidth and that of the others on your host, if you saturate your incoming bandwidth easily it might be worth limiting your download speed to ~80% because saturating downstream can effectively throttle upstream (and vice versa) - meaning that while your client is initially downloading it is not able to serve much in the other direction until it is finished (and demand drops because most others have finished too).
Also on the machine with much slower IO try limiting the number of concurrent streams. Though if your neighbours are pretty noisy then neither measure will help a lot. Another way to get 1:1 or better is to stay in the swarm longer to catch the tail or later downloaders, though that requires much more space to keep old content lying around.
For that, go with the one that has the better bandwidth and lower latency to your usual location(s).
keep the one with less iowait
If the OpenVZ one is from Hostsolutions you might want to consider asking support for upgrade to KVM.