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.
How is ping to Clouvider this low on Vultr?
So I got a trial of Vultr and decided to run YABS.
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
# Yet-Another-Bench-Script #
# v2022-11-22 #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Wed 30 Nov 2022 02:41:45 PM UTC
Basic System Information:
---------------------------------
Uptime : 0 days, 21 hours, 5 minutes
Processor : Intel Core Processor (Broadwell, no TSX, IBRS)
CPU cores : 1 @ 2394.454 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 976.6 MiB
Swap : 2.3 GiB
Disk : 23.4 GiB
Distro : Debian GNU/Linux 11 (bullseye)
Kernel : 5.10.0-18-amd64
fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k (IOPS) | 64k (IOPS)
------ | --- ---- | ---- ----
Read | 92.50 MB/s (23.1k) | 166.20 MB/s (2.5k)
Write | 92.74 MB/s (23.1k) | 167.07 MB/s (2.6k)
Total | 185.25 MB/s (46.3k) | 333.27 MB/s (5.2k)
| |
Block Size | 512k (IOPS) | 1m (IOPS)
------ | --- ---- | ---- ----
Read | 182.61 MB/s (356) | 170.81 MB/s (166)
Write | 192.31 MB/s (375) | 182.18 MB/s (177)
Total | 374.92 MB/s (731) | 352.99 MB/s (343)
iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed | Ping
----- | ----- | ---- | ---- | ----
Clouvider | London, UK (10G) | 8.36 Gbits/sec | 8.24 Gbits/sec | 0.687 ms
Scaleway | Paris, FR (10G) | 7.78 Gbits/sec | busy | 8.19 ms
NovoServe | North Holland, NL (40G) | 6.45 Gbits/sec | 8.32 Gbits/sec | 7.46 ms
Uztelecom | Tashkent, UZ (10G) | 2.22 Gbits/sec | 3.95 Gbits/sec | 81.1 ms
Clouvider | NYC, NY, US (10G) | 2.63 Gbits/sec | 2.97 Gbits/sec | 75.0 ms
Clouvider | Dallas, TX, US (10G) | 1.83 Gbits/sec | 2.05 Gbits/sec | 110 ms
Clouvider | Los Angeles, CA, US (10G) | 1.37 Gbits/sec | 1.54 Gbits/sec | 130 ms
iperf3 Network Speed Tests (IPv6):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed | Ping
----- | ----- | ---- | ---- | ----
Clouvider | London, UK (10G) | 8.12 Gbits/sec | 6.89 Gbits/sec | 0.718 ms
Scaleway | Paris, FR (10G) | busy | busy | 7.49 ms
NovoServe | North Holland, NL (40G) | 6.45 Gbits/sec | 8.04 Gbits/sec | 7.38 ms
Uztelecom | Tashkent, UZ (10G) | 2.60 Gbits/sec | 3.23 Gbits/sec | 80.9 ms
Clouvider | NYC, NY, US (10G) | 2.60 Gbits/sec | 2.88 Gbits/sec | 74.9 ms
Clouvider | Dallas, TX, US (10G) | 1.65 Gbits/sec | 1.64 Gbits/sec | 110 ms
Clouvider | Los Angeles, CA, US (10G) | 1.34 Gbits/sec | 1.47 Gbits/sec | 130 ms
Geekbench 5 Benchmark Test:
---------------------------------
Test | Value
|
Single Core | 701
Multi Core | 700
Full Test | https://browser.geekbench.com/v5/cpu/18986481
What? ~0.7ms ping to Clouvider?
Can anyone explain me what the fuck is going on lmao
Comments
Your Vultr is in the UK as well?
They're located in the same data center? Otherwise it can't be this low.
lmao
Yeah. London location.
Are they?
lmao
I don't know, but usually, even if they're in the same city, the latency still can't be lower than 1ms. So I guess or assume they might be in the same data center, or in two very close data centers.
Non-zero chance. Who knows?
Ping from Clouvider Frankfurt to Vultr Frankfurt
lmao
waaa
I've only seen pings that low between systems with the same provider (e.g., on the same switch).
Even within a single datacenter, but across two different providers with their own networks, I frequently see higher latency than that.
With these numbers, I would suspect Clouvider may be a colocation client of Vultr/Constant.
Pings routed through quantum tunneling?
Ummm...
Clouvider hosted on same rack?
This is definitely possible! We have a link across Amsterdam traversing three datacentres on different campuses across the city, with a RTT of 0.7ms.
That’s highly possible (Vultr rents rack from DCs except for New Jersey), or their datacenters are close to each other and directly connected.
Lmao ok
lmfao
Use your real money @Otus9051
Wow that is indeed really low and like others said as if it's from the same DC.
@Clouvider
We are not Customers of Vultr in any location. They may be using the same highly interconnected building as we do.
Run an mtr - everything is there.
It's coming from inside the house!
Here's MTR
Basically sounds like they're either in the same datacenter or Vultr probably uses Arelion's network or both.
This is possible and it's nothing too surprisingly. Other who say this is impossible is misinformed.
Looking at the rDNS, they most likely are NOT within the same datacenter, but they are common clients of 1299 - just the Telia network being speedy speedy :-)
Could be within the same building but different floors or spaces.
It's like 0.3-0.4ms when they are in the same facility.