Free DirectAdmin. $5 VPS: 2GB RAM, 20GB SSD, Dallas, Amsterdam, London, Fremont, Bucharest

Andru

Any ETAfor Cluj? Cant wait to test there 😍

intovps

@zuby2402 said:
Any ETAfor Cluj? Cant wait to test there 😍

About 2 weeks

sin

Does this support custom iso?

fpmagic

@zuby2402 said:
Any ETAfor Cluj? Cant wait to test there 😍

Interesting, is there enough connectivity and localized peering to make this beneficial over the capital? We faced a similar situation in the UK where by any northern hosting went via London anyway but that's changed over last five years with growing peering fabrics in Manchester and even Leeds. Unrelated, what's the co-working scene like in Cluj? Romania is on my list before the winter

Andru

I live near Cluj...

emgh

@zuby2402 said:
I live near Cluj...

I don't...

alexvolk

@intovps said:
That's something we might do. I'm curious, are you actually making more than 1TB/2TB or just want to be on the same side?

Of course not! It's always amazing to have extra premium traffic just in case

intovps

@alexvolk said:

@intovps said:
That's something we might do. I'm curious, are you actually making more than 1TB/2TB or just want to be on the same side?

Of course not! It's always amazing to have extra premium traffic just in case

That's useful feedback. I'll keep that in mind. Thanks!

intovps

@zuby2402 said:
I live near Cluj...

First batch of hardware has been racked in Cluj-Napoca a month ago.

We've been working to get every possible bit of performance out the the 40G network cards and from the 40G fiberswitches. Drivers, testing optimal OpenStack network setup, active-active setup that can push close to 80Gbs etc.

Big SATA storage as block device (OpenStack Cinder) and object storage (a la S3, via OpenStack swift) is next in line.

New toys lot of fun!

See more pictures here: https://hosterion.ro/blog/in-curand-servicii-de-gazduire-web-in-cluj-napoca

intovps

@sin said:
Does this support custom iso?

Custom ISO upload feature is already available in Fleio.

We've been considering some security and performance implication regarding enabling ISO upload.

But I think we're fine. I'll talk to our staff and see if it's time we enable it on IntoVPS.

intovps

@fpmagic said:

@zuby2402 said:
Any ETAfor Cluj? Cant wait to test there 😍

Interesting, is there enough connectivity and localized peering to make this beneficial over the capital? We faced a similar situation in the UK where by any northern hosting went via London anyway but that's changed over last five years with growing peering fabrics in Manchester and even Leeds. Unrelated, what's the co-working scene like in Cluj? Romania is on my list before the winter

In Cluj we're in GTS, a non-carrier-neutral-friendly data center, but never the less, a reputable premium ISP (actually one of the most expensive, if not THE), part of Deutsche Telekom. Peering is also done mainly through Bucharest in Romania (we have some plans to change that, as a side note ).

Even so, Cluj gives a few advantages like disaster recovery location for Bucharest located data (different tectonic plate, less seismic risk), direct connection for local IT companies (which are plenty - https://techcrunch.com/2016/04/06/the-silicon-valley-of-transylvania/ ), exotic location for international customers and some amazing deals on the way.

Cluj-Napoca has some really good co-working facilities, some that I know personally:

• https://clujhub.ro/
• http://www.siliconforest.ro/

I see some more on Google.

PM me and come by our office for an espresso.

MikePT

Damn. I am starting to love your services. Even more than previously. You guys rock. I did manage a few vps on intovps many years ago though I never bought one for myself... It looks like its time to.

MechanicWeb

You guys are humble. Why don't you include in your offerings that you are the creator of intoDNS? When I found out about that fact, it was a pleasant surprise and we are grateful to you for creating such an amazing tool.

Warm regards from the deepest of our hearts.

intovps

@MechanicWeb said:
You guys are humble. Why don't you include in your offerings that you are the creator of intoDNS? When I found out about that fact, it was a pleasant surprise and we are grateful to you for creating such an amazing tool.

Warm regards from the deepest of our hearts.

We do brag about it on our about page and in the IntoDNS footer.

Cheers!

intovps

Testing our upcoming Cluj location on NVMe vs. Amsterdam on SSD's

Cluj-Napoca, GTS data center on NVMe

wget -qO- bench.sh | bash

----------------------------------------------------------------------
CPU model            : Intel Xeon Processor (Skylake, IBRS)
Number of cores      : 2
CPU frequency        : 2099.998 MHz
Total size of Disk   : 58.1 GB (1.2 GB Used)
Total amount of Mem  : 7976 MB (97 MB Used)
Total amount of Swap : 0 MB (0 MB Used)
System uptime        : 0 days, 0 hour 6 min
Load average         : 0.00, 0.02, 0.00
OS                   : Ubuntu 18.04.2 LTS
Arch                 : x86_64 (64 Bit)
Kernel               : 4.15.0-54-generic
----------------------------------------------------------------------
I/O speed(1st run)   : 947 MB/s
I/O speed(2nd run)   : 1000 MB/s
I/O speed(3rd run)   : 940 MB/s
Average I/O speed    : 962.3 MB/s
----------------------------------------------------------------------
Node Name                       IPv4 address            Download Speed
CacheFly                        205.234.175.175         72.1MB/s      
Linode, Tokyo, JP               106.187.96.148          7.54MB/s      
Linode, Singapore, SG           139.162.23.4            8.06MB/s      
Linode, London, UK              176.58.107.39           49.0MB/s      
Linode, Frankfurt, DE           139.162.130.8           73.8MB/s      
Linode, Fremont, CA             50.116.14.9             12.2MB/s      
Softlayer, Dallas, TX           173.192.68.18           14.8MB/s      
Softlayer, Seattle, WA          67.228.112.250          12.7MB/s      
Softlayer, Frankfurt, DE        159.122.69.4            40.1MB/s      
Softlayer, Singapore, SG        119.81.28.170           11.3MB/s      
Softlayer, HongKong, CN         119.81.130.170          914KB/s       
----------------------------------------------------------------------
Node Name                       IPv6 address            Download Speed
Linode, Atlanta, GA             2600:3c02::4b           13.8MB/s      
Linode, Dallas, TX              2600:3c00::4b           15.8MB/s      
Linode, Newark, NJ              2600:3c03::4b           3.84MB/s      
Linode, Singapore, SG           2400:8901::4b           12.0MB/s      
Linode, Tokyo, JP               2400:8900::4b           7.73MB/s      
Softlayer, San Jose, CA         2607:f0d0:2601:2a::4    8.68MB/s 
Softlayer, Paris, FR            2a03:8180:1301:8::4     48.6MB/s      
Softlayer, Singapore, SG        2401:c900:1101:8::2     11.7MB/s      
Softlayer, Tokyo, JP            2401:c900:1001:16::4    7.15MB/s      

Amsterdam (Leaseweb in Evoswitch data center) on SSD:

wget -qO- bench.sh | bash

----------------------------------------------------------------------
CPU model            : Intel Core Processor (Skylake, IBRS)
Number of cores      : 2
CPU frequency        : 2399.988 MHz
Total size of Disk   : 58.1 GB (5.4 GB Used)
Total amount of Mem  : 7976 MB (1931 MB Used)
Total amount of Swap : 0 MB (0 MB Used)
System uptime        : 7 days, 18 hour 58 min
Load average         : 0.48, 0.21, 0.12
OS                   : Ubuntu 18.04.2 LTS
Arch                 : x86_64 (64 Bit)
Kernel               : 4.15.0-54-generic
----------------------------------------------------------------------
I/O speed(1st run)   : 434 MB/s
I/O speed(2nd run)   : 469 MB/s
I/O speed(3rd run)   : 473 MB/s
Average I/O speed    : 458.7 MB/s
----------------------------------------------------------------------
Node Name                       IPv4 address            Download Speed
CacheFly                        205.234.175.175         60.1MB/s      
Linode, Tokyo, JP               106.187.96.148          9.30MB/s      
Linode, Singapore, SG           139.162.23.4            4.33MB/s      
Linode, London, UK              176.58.107.39           43.8MB/s      
Linode, Frankfurt, DE           139.162.130.8           90.7MB/s      
Linode, Fremont, CA             50.116.14.9             5.52MB/s      
Softlayer, Dallas, TX           173.192.68.18           10.9MB/s      
Softlayer, Seattle, WA          67.228.112.250          10.1MB/s      
Softlayer, Frankfurt, DE        159.122.69.4            48.3MB/s      
Softlayer, Singapore, SG        119.81.28.170           6.59MB/s      
Softlayer, HongKong, CN         119.81.130.170          3.91MB/s      
----------------------------------------------------------------------

MikePT

Could you please run nench.sh?

intovps

@MikePT said:
Could you please run nench.sh?

root@cluj:~# (wget -qO- wget.racing/nench.sh | bash; wget -qO- wget.racing/nench.sh | bash) 2>&1 | tee nench.log
-------------------------------------------------
 nench.sh v2019.06.29 -- https://git.io/nench.sh
 benchmark timestamp:    2019-07-17 09:17:25 UTC
-------------------------------------------------

Processor:    Intel Xeon Processor (Skylake, IBRS)
CPU cores:    2
Frequency:    2099.998 MHz
RAM:          7.8G
Swap:         -
Kernel:       Linux 4.15.0-54-generic x86_64

Disks:
sda     60G  HDD

CPU: SHA256-hashing 500 MB
    4.004 seconds
CPU: bzip2-compressing 500 MB
    6.809 seconds
CPU: AES-encrypting 500 MB
    1.537 seconds

ioping: seek rate
    min/avg/max/mdev = 65.3 us / 138.9 us / 7.09 ms / 45.4 us
ioping: sequential read speed
    generated 24.6 k requests in 5.00 s, 6.01 GiB, 4.92 k iops, 1.20 GiB/s

dd: sequential write speed
    1st run:    953.67 MiB/s
    2nd run:    953.67 MiB/s
    3rd run:    885.96 MiB/s
    average:    931.10 MiB/s

IPv4 speedtests
    your IPv4:    185.250.105.xxxx

    Cachefly CDN:         70.82 MiB/s
    Leaseweb (NL):        56.56 MiB/s
    Softlayer DAL (US):   13.99 MiB/s
    Online.net (FR):      55.58 MiB/s
    OVH BHS (CA):         19.67 MiB/s

IPv6 speedtests
    your IPv6:    2a06:cd40:400:xxxx

    Leaseweb (NL):        31.54 MiB/s
    Softlayer DAL (US):   0.00 MiB/s
    Online.net (FR):      31.55 MiB/s
    OVH BHS (CA):         13.28 MiB/s
-------------------------------------------------

-------------------------------------------------
 nench.sh v2019.06.29 -- https://git.io/nench.sh
 benchmark timestamp:    2019-07-17 09:18:34 UTC
-------------------------------------------------

Processor:    Intel Xeon Processor (Skylake, IBRS)
CPU cores:    2
Frequency:    2099.998 MHz
RAM:          7.8G
Swap:         -
Kernel:       Linux 4.15.0-54-generic x86_64

Disks:
sda     60G  HDD

CPU: SHA256-hashing 500 MB
    4.008 seconds
CPU: bzip2-compressing 500 MB
    6.803 seconds
CPU: AES-encrypting 500 MB
    1.463 seconds

ioping: seek rate
    min/avg/max/mdev = 65.4 us / 134.5 us / 7.97 ms / 76.5 us
ioping: sequential read speed
    generated 22.9 k requests in 5.00 s, 5.58 GiB, 4.57 k iops, 1.12 GiB/s

dd: sequential write speed
    1st run:    934.60 MiB/s
    2nd run:    953.67 MiB/s
    3rd run:    953.67 MiB/s
    average:    947.32 MiB/s

IPv4 speedtests
    your IPv4:    185.250.105.xxxx

    Cachefly CDN:         71.89 MiB/s
    Leaseweb (NL):        60.44 MiB/s
    Softlayer DAL (US):   14.49 MiB/s
    Online.net (FR):      61.97 MiB/s
    OVH BHS (CA):         20.17 MiB/s

IPv6 speedtests
    your IPv6:    2a06:cd40:400:xxxx

    Leaseweb (NL):        56.73 MiB/s
    Softlayer DAL (US):   0.00 MiB/s
    Online.net (FR):      48.81 MiB/s
    OVH BHS (CA):         18.55 MiB/s
-------------------------------------------------

MikePT

That seems much better than my experience @intovps. Love the speed and IOPS!

TimboJones

@alexvolk said:

@intovps said:
As you know, we can do that right now.

But those are shared cores and that will overload the servers, making those two cores worth about the same like 1 core.

Another reason is that we have disabled Hyperthreading on all our servers to mitigate the Intel Xeon vulnerabilities. We're also running the latest kernel. Just best security practice in any multi-tenant environment. I hope other hosters have disabled HT, as well.

Improved security lost us about 30% of compute power.

I've been using your services from 2011.

I think with recent additions of big players you can't simply offer the similar looking specs at the higher price. You can do that but don't expect someone from here would prefer you.

I'm not sure who cares about your lost computer power except yourself?

Speak for yourself, that is useful info and suitable for a hosting forum discussion board. This is hands on experience. I'm sure your life was negatively affected by reading one sentence that you had to be rude for.

alexvolk

@TimboJones said:
Speak for yourself, that is useful info and suitable for a hosting forum discussion board. This is hands on experience. I'm sure your life was negatively affected by reading one sentence that you had to be rude for.

I'm not trying to be specially rude

If provider wants to compete with others then I really don't care if they've lost some computing power due to some vulnerability.

TimboJones

@alexvolk said:

@TimboJones said:
Speak for yourself, that is useful info and suitable for a hosting forum discussion board. This is hands on experience. I'm sure your life was negatively affected by reading one sentence that you had to be rude for.

I'm not trying to be specially rude

If provider wants to compete with others then I really don't care if they've lost some computing power due to some vulnerability.

OK. Just letting you know that IS useful information, as it is a security improvement that many others DO care about, as well with Ryzen update this month and more providers setting up AMD boxes instead of Xeon boxes. No exploits needing to disable HT equivalent on AMD CPUs.