Big Power Hosting Review – Miami KVM VPS Hosting

I found a really attractive offer online from a company called Big Power Hosting who offered the following in Miami:

*exceed 2 TB/mo then throttled back to 10Mbps.

Don’t like Miami?

Big Power Hosting has servers in New York (NJ) and Los Angeles for that same offer.

I thought to myself that’s a pretty attractive offer and even more attractive with the 60off coupon being valid until the end of 2018 which pushes it into a lot of pricing that Vultr, Digital Ocean, Linode, and other “cloud” providers offer.

The CPU on the VPS node is a Intel Xeon E5-1650 v4 @ 3.6 Ghz and 15mb cache.

The network is amazing. I think the node I’m on is idling but I’m using close to full download on the 1 Gbps uplink where I downloaded the 100mb test file at 112M/s in less than 1 second.

CPU information:

# cat /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 79
model name : Intel(R) Xeon(R) CPU E5-1650 v4 @ 3.60GHz
stepping : 1
microcode : 0x1
cpu MHz : 3599.998
cache size : 15360 KB
fpu : yes
fpu_exception : yes
cpuid level : 13
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon rep_good nopl xtopology pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch arat xsaveopt fsgsbase smep erms
bogomips : 7199.99
clflush size : 64
cache_alignment : 64
address sizes : 46 bits physical, 48 bits virtual
power management:

processor : 1
vendor_id : GenuineIntel
cpu family : 6
model : 79
model name : Intel(R) Xeon(R) CPU E5-1650 v4 @ 3.60GHz
stepping : 1
microcode : 0x1
cpu MHz : 3599.998
cache size : 15360 KB
fpu : yes
fpu_exception : yes
cpuid level : 13
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon rep_good nopl xtopology pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch arat xsaveopt fsgsbase smep erms
bogomips : 7199.99
clflush size : 64
cache_alignment : 64
address sizes : 46 bits physical, 48 bits virtual
power management:

Here is a hardware benchmark and some network benchmarking but I don’t like the small amount of network testing available in this script because you could reasonably assume that it’s not an optimal connection when it’s not exactly a comprehensive network test of multiple sources where you’d download 10 or more test files for more data.

-------------------------------------------------
 nench.sh v2018.04.14 -- https://git.io/nench.sh
 benchmark timestamp:    2018-11-16 16:57:53 UTC
-------------------------------------------------

Processor:    Intel(R) Xeon(R) CPU E5-1650 v4 @ 3.60GHz
CPU cores:    2
Frequency:    3599.998 MHz
RAM:          496M
Swap:         1.0G
Kernel:       Linux 3.2.0-4-amd64 x86_64

Disks:
sda     20G  HDD

CPU: SHA256-hashing 500 MB
    2.225 seconds
CPU: bzip2-compressing 500 MB
    4.057 seconds
CPU: AES-encrypting 500 MB
    1.081 seconds

ioping: seek rate
    min/avg/max/mdev = 144.6 us / 158.4 us / 982.3 us / 23.7 us
ioping: sequential read speed
    generated 20.2 k requests in 5.00 s, 4.94 GiB, 4.05 k iops, 1012.0 MiB/s

dd: sequential write speed
    1st run:    287.06 MiB/s
    2nd run:    286.10 MiB/s
    3rd run:    276.57 MiB/s
    average:    283.24 MiB/s

IPv4 speedtests
    your IPv4:    103.195.102.xxxx

    Cachefly CDN:         103.93 MiB/s
    Leaseweb (NL):        2.12 MiB/s
    Softlayer DAL (US):   0.00 MiB/s
    Online.net (FR):      2.13 MiB/s
    OVH BHS (CA):         5.52 MiB/s

IPv6 speedtests
    your IPv6:    2605:9880:400:xxxx

    Leaseweb (NL):        2.04 MiB/s
    Softlayer DAL (US):   0.00 MiB/s
    Online.net (FR):      2.15 MiB/s
    OVH BHS (CA):         5.42 MiB/s
-------------------------------------------------

For example, when I use bench.sh instead of nench.sh

# wget -qO- bench.sh | bash
----------------------------------------------------------------------
CPU model            : Intel(R) Xeon(R) CPU E5-1650 v4 @ 3.60GHz
Number of cores      : 2
CPU frequency        : 3599.998 MHz
Total size of Disk   : 38.0 GB (2.6 GB Used)
Total amount of Mem  : 496 MB (74 MB Used)
Total amount of Swap : 1022 MB (0 MB Used)
System uptime        : 0 days, 15 hour 25 min
Load average         : 0.03, 0.10, 0.04
OS                   : Debian GNU/Linux 7
Arch                 : x86_64 (64 Bit)
Kernel               : 3.2.0-4-amd64
----------------------------------------------------------------------
I/O speed(1st run)   : 285 MB/s
I/O speed(2nd run)   : 271 MB/s
I/O speed(3rd run)   : 277 MB/s
Average I/O speed    : 277.7 MB/s
----------------------------------------------------------------------
Node Name                       IPv4 address            Download Speed
CacheFly                        205.234.175.175         111MB/s
Linode, Tokyo, JP               106.187.96.148          1.43MB/s
Linode, Singapore, SG           139.162.23.4            942KB/s
Linode, London, UK              176.58.107.39           245KB/s
Linode, Frankfurt, DE           139.162.130.8           2.01MB/s
Linode, Fremont, CA             50.116.14.9             2.22MB/s
Softlayer, Dallas, TX           173.192.68.18           7.63MB/s
Softlayer, Seattle, WA          67.228.112.250          3.38MB/s
Softlayer, Frankfurt, DE        159.122.69.4            2.07MB/s
Softlayer, Singapore, SG        119.81.28.170           907KB/s
Softlayer, HongKong, CN         119.81.130.170          1.00MB/s
----------------------------------------------------------------------
Node Name                       IPv6 address            Download Speed
Linode, Atlanta, GA             2600:3c02::4b           11.8MB/s
Linode, Dallas, TX              2600:3c00::4b           7.22MB/s
Linode, Newark, NJ              2600:3c03::4b           6.47MB/s
Linode, Singapore, SG           2400:8901::4b           1007KB/s
Linode, Tokyo, JP               2400:8900::4b           1.31MB/s
Softlayer, San Jose, CA         2607:f0d0:2601:2a::4    3.26MB/s
Softlayer, Washington, WA       2607:f0d0:3001:78::2    7.02MB/s
Softlayer, Paris, FR            2a03:8180:1301:8::4     1.72MB/s
Softlayer, Singapore, SG        2401:c900:1101:8::2     975KB/s
Softlayer, Tokyo, JP            2401:c900:1001:16::4    1.39MB/s
----------------------------------------------------------------------

I know in some examples its not looking good but nothing against this company but Miami is a location where you deploy in Miami if Miami is a requirement for Florida or Southeast US users. I know in Florida, three points where traffic come in and out of the state are in Jacksonville, Orlando and Miami where even in Jacksonville traffic can be routed down to Miami then back up to Atlanta.

Another group of users is Central and South America because Brazil has good connectivity to Miami due to underwater fiber optic connections. But your users in Central or West US aren’t going to have the exact best connectivity to the very very end of the southern tip of Florida.

Conclusion

So far so good with Big Power Hosting.

I opened up a ticket with the sales department asking about a particular type of VPS hosting plan that is becoming popular. Brian and Jason have been keeping me updated about it and once they have more information, which supposedly is supposed to happen this weekend, I will post it on here.

OneProvider VPS Review – New York City

I’ve been using a OneProvider OneCloud VPS for the past 30 days in their New York City location.

OneProvider is an affordable dedicated server hosting provider with locations across North and South America, Europe, Asia and even Australia. In the multiple locations they provide, their VPS solution called “OneCloud” is also offered.

Plans for OneCloud start out at $5 per month for 512mb RAM, 1 CPU, 20GB SSD disk space, 1 IPv4 address and 1TB transfer on a 1Gbps port. OneProvider was running a 50% off in North America deal so I grabbed a New York City and Chicago service for $2.50/mo in each location and looking at the OnePanel it appears this offer was for a limited time which is unavailable.

There are over 25 data centers to choose from with OneProvider with this offer with locations such as:

  • Atlanta, US
  • Chicago, US
  • Los Angeles, US
  • Mexico City, MX
  • Montreal, CA
  • New York, US
  • Vancouver, CA
  • Amsterdam, NL
  • Bucharest, RO
  • Copenhagen, DK
  • Frankfurt, DE
  • London, UK
  • Milan, IT
  • Moscow, RU
  • Oslo, NO
  • Paris, FR
  • Stockholm, SE
  • Valencia, ES
  • Vienna, AT
  • Zurich, CH
  • Hong Kong, HK
  • Singapore, SG
  • Tokyo, JP
  • Melbourne, AU
  • Johannesburg, ZA

As you can see above, more than enough choice and OneProvider offers 2 months free if you prepay for an entire year. The stability at the New York City location is more than reliable and I’ve been extremely pleased with the stability of the location with both the power and the network.


OneProvider OneCloud Review:

I purchased my OneProvider OneCloud New York VPS about 34 days ago and installed snmp monitoring on it so I could add it to my LibreNMS installation to monitor the network stability. There are a few small gaps in the reporting but it was not enough of a gap/lapse for LibreNMS to trigger this as a downtime alert so I don’t really consider this “downtime”.

When you sign up with OneProvider, you have to submit a “photo ID selfie” with you holding up your photo ID and taking a selfie with a smart phone, tablet or webcam. I found it slightly unusual but if its what OneProvider needs to do to keep the fraud and abuse down with their service to continue to offer their pricing which is extremely generous, I don’t have an issue as it’s their own policy.

What I like and find interesting about OneProvider’s billing for their VPS services is that it’s pro-rated from the first of the month. Nothing is more confusing than when 10+ VPS services are due and having to keep up with it. The 2 services I ordered were pro-rated to the first of the month when I ordered and had the monthly recurring rate added to the invoice. Some customers may not like it but I like it because of the ease of keeping up with knowing that your invoice is due at the first of the month, every month, and not on the end of your 30 day usage of the service.

The virtualization is QEMU based, not Xen or OpenVZ like other providers and not KVM based like OVH VPS product line. I don’t have an issue with this and haven’t noticed a difference between KVM, OpenVZ or QEMU. If it works, it works and I’m happy. When it doesn’t work or doesn’t work like its supposed to is when I have an issue.

Network and I/O benchmarks ran on April 25th 2016 at 0930 EST
Network and I/O benchmarks ran on April 25th 2016 at 0930 EST

The service is connected on a 1Gbps port, not a 100Mbps port like with OVH VPS, so the network connectivity is a little bit faster. A lot of people have been curious about the network locations since OneProvider doesn’t provide much details but the New York City facility provider is Ubiquity Servers/NobisTech according to the IP assignment and the traceroutes.

I don’t have an issue with this as it’s reliable and I’ve never had any outages related to this service so there’s no need to make an issue out of this non-issue.

My guess is OneProvider uses its purchasing power with facility providers like Ubiquity Servers/NobisTech to negotiate the best deals which benefit us, the customers.

The network benchmarks, ran at 0948 EST, aren’t looking that great. The connectivity to Cachefly’s CDN service is excellent but connectivity to Colo@, Softlayer and Linode aren’t looking too great. This is when OneProvider running their own network at Ubiquity/NobisTech would be more advantageous rather than OneProvider being reliant on Ubiquity/NobisTech fixing the network connectivity issue during a period of the network being slow.

Conclusion:

With the limited time 50% off per month offer now unavailable, I still feel that this offer is an attractive offer especially to customers who need a variety of options for location with similar resource requirements.

I think Ubiquity Server/NobisTech’s network quality is the only disadvantage for OneProvider’s OneCloud platform. The pricing and resources are right, just the shortcoming of the network facility is what could harm the reputation of the platform. If OneProvider moves around too much in establishing a better location, they run the risk I feel of having the service interpreted as being not the greatest which would harm the reliability a hosting provider would try to project to their current and future customers.

However it could be worse such as having the platform reliant on ColoCrossing’s Buffalo, NY location. I will give OneProvider credit for actually selling something in the New York City area and not selling Buffalo while claiming it’s New York City even with the “it’s less than 10ms difference” when Buffalo is on the northwest corner of New York state and New York City is over 350 physical miles away on the other side of the state but in the southeast corner.

I look forward in trying OneProvider’s other locations once discounts and coupon codes come out to make it under $5 per month.