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.
HostCram offers 5-10x the performance/value compared to Vultr HF
I just moved from a $12/month 1vCPU, 2GB RAM Vultr HF plan to the $7/month KVM-3 plan from HostCram (via this thread's deals and coupons).
I have to say that I'm extremely impressed so far. Here are some comparisons
YABS - Hostcram Disk 5x faster, CPU Single Core 2x, MultiCore 5.5x faster
HostCram
root@ubuntu:~# wget -qO- yabs.sh | bash
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
# Yet-Another-Bench-Script #
# v2021-06-05 #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Sun 26 Sep 2021 04:09:33 PM UTC
Basic System Information:
---------------------------------
Processor : 11th Gen Intel(R) Core(TM) i9-11900K @ 3.50GHz
CPU cores : 3 @ 3504.000 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 2.9 GiB
Swap : 2.9 GiB
Disk : 68.4 GiB
fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k (IOPS) | 64k (IOPS)
------ | --- ---- | ---- ----
Read | 492.10 MB/s (123.0k) | 3.92 GB/s (61.3k)
Write | 493.40 MB/s (123.3k) | 3.94 GB/s (61.6k)
Total | 985.50 MB/s (246.3k) | 7.86 GB/s (122.9k)
| |
Block Size | 512k (IOPS) | 1m (IOPS)
------ | --- ---- | ---- ----
Read | 4.06 GB/s (7.9k) | 3.89 GB/s (3.8k)
Write | 4.28 GB/s (8.3k) | 4.15 GB/s (4.0k)
Total | 8.35 GB/s (16.3k) | 8.05 GB/s (7.8k)
iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed
| | |
Clouvider | London, UK (10G) | 830 Mbits/sec | 150 Mbits/sec
Online.net | Paris, FR (10G) | 788 Mbits/sec | 117 Mbits/sec
WorldStream | The Netherlands (10G) | 778 Mbits/sec | 181 Mbits/sec
Biznet | Jakarta, Indonesia (1G) | busy | busy
Clouvider | NYC, NY, US (10G) | 838 Mbits/sec | 555 Mbits/sec
Velocity Online | Tallahassee, FL, US (10G) | 828 Mbits/sec | 733 Mbits/sec
Clouvider | Los Angeles, CA, US (10G) | 820 Mbits/sec | 445 Mbits/sec
Iveloz Telecom | Sao Paulo, BR (2G) | 516 Mbits/sec | 97.9 Mbits/sec
Geekbench 5 Benchmark Test:
---------------------------------
Test | Value
|
Single Core | 1813
Multi Core | 4577
Full Test | https://browser.geekbench.com/v5/cpu/10087360
Vultr
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
# Yet-Another-Bench-Script #
# v2021-06-05 #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Fri 24 Sep 2021 05:00:03 PM UTC
Basic System Information:
---------------------------------
Processor : Intel Core Processor (Skylake, IBRS)
CPU cores : 1 @ 3695.996 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 1.9 GiB
Swap : 2.0 GiB
Disk : 60.0 GiB
fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k (IOPS) | 64k (IOPS)
------ | --- ---- | ---- ----
Read | 332.28 MB/s (83.0k) | 604.69 MB/s (9.4k)
Write | 333.16 MB/s (83.2k) | 607.87 MB/s (9.4k)
Total | 665.44 MB/s (166.3k) | 1.21 GB/s (18.9k)
| |
Block Size | 512k (IOPS) | 1m (IOPS)
------ | --- ---- | ---- ----
Read | 659.20 MB/s (1.2k) | 705.93 MB/s (689)
Write | 694.23 MB/s (1.3k) | 752.95 MB/s (735)
Total | 1.35 GB/s (2.6k) | 1.45 GB/s (1.4k)
iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed
| | |
Clouvider | London, UK (10G) | 1.39 Gbits/sec | 1.32 Gbits/sec
Online.net | Paris, FR (10G) | 1.41 Gbits/sec | 1.27 Gbits/sec
WorldStream | The Netherlands (10G) | busy | busy
Biznet | Jakarta, Indonesia (1G) | busy | busy
Clouvider | NYC, NY, US (10G) | 2.79 Gbits/sec | 2.52 Gbits/sec
Velocity Online | Tallahassee, FL, US (10G) | 2.49 Gbits/sec | 2.34 Gbits/sec
Clouvider | Los Angeles, CA, US (10G) | 3.28 Gbits/sec | 4.29 Gbits/sec
Iveloz Telecom | Sao Paulo, BR (2G) | 743 Mbits/sec | 928 Mbits/sec
Geekbench 5 Benchmark Test:
---------------------------------
Test | Value
|
Single Core | 921
Multi Core | 827
Full Test | https://browser.geekbench.com/v5/cpu/10042351
Speedtests (the slow iperf3 receive speeds for HostCram seemed suspicious, so I ran a variety directly from speedtest.net. They seem to match the send speeds now. They're considerably slower than the Vultr 1.4-4.3 Ghz speeds, but should be more than fine for my needs.
Server: Swish Fibre - London (id = 34948)
ISP: HostCram LLC
Latency: 106.97 ms (0.18 ms jitter)
Download: 950.64 Mbps (data used: 1.6 GB)
Upload: 754.06 Mbps (data used: 1.2 GB)
Packet Loss: 0.0%
Result URL: https://www.speedtest.net/result/c/7d6255f8-d2b5-4895-a7ee-3e8272478459
root@ubuntu:~# speedtest --server-id=24215
Speedtest by Ookla
Server: ORANGE FRANCE - Paris (id = 24215)
ISP: HostCram LLC
Latency: 111.45 ms (0.05 ms jitter)
Download: 597.22 Mbps (data used: 810.4 MB)
Upload: 735.08 Mbps (data used: 1.0 GB)
Packet Loss: 0.0%
Result URL: https://www.speedtest.net/result/c/c986e597-8273-46a3-98a0-003a712ff250
root@ubuntu:~# speedtest --server-id=44747
Speedtest by Ookla
Server: ARC-France - Paris (id = 44747)
ISP: HostCram LLC
Latency: 113.77 ms (0.15 ms jitter)
Download: 502.44 Mbps (data used: 583.3 MB)
Upload: 628.16 Mbps (data used: 1.1 GB)
Packet Loss: 0.0%
Result URL: https://www.speedtest.net/result/c/d4ea3aef-b707-446c-a142-e6cff18dd9bb
root@ubuntu:~# speedtest --server-id=6554
Speedtest by Ookla
Server: WorldStream B.V. - Naaldwijk (id = 6554)
ISP: HostCram LLC
Latency: 113.88 ms (0.29 ms jitter)
Download: 515.44 Mbps (data used: 838.0 MB)
Upload: 735.45 Mbps (data used: 995.1 MB)
Packet Loss: 0.0%
Result URL: https://www.speedtest.net/result/c/3051af6c-7e72-470b-a03d-0177e7a5f5cb
root@ubuntu:~# speedtest --server-id=40095
Speedtest by Ookla
Server: Claro Guatemala (100G) - Guatemala City (id = 40095)
ISP: HostCram LLC
Latency: 59.48 ms (0.06 ms jitter)
Download: 840.05 Mbps (data used: 1.3 GB)
Upload: 874.25 Mbps (data used: 1.1 GB)
Packet Loss: 0.0%
Result URL: https://www.speedtest.net/result/c/00bb8da9-2a68-4b20-95a4-cc1ca83cd611
root@ubuntu:~# speedtest --server-id=3455
Speedtest by Ookla
Server: Claro Perú - Lima (id = 3455)
ISP: HostCram LLC
Latency: 93.75 ms (0.04 ms jitter)
Download: 792.71 Mbps (data used: 1.1 GB)
Upload: 808.02 Mbps (data used: 1.1 GB)
Packet Loss: 0.0%
Result URL: https://www.speedtest.net/result/c/85bfd5ba-beca-4427-90c4-b6ca01f8be7f
root@ubuntu:~# speedtest --server-id=35056
Speedtest by Ookla
Server: Clouvider Ltd - Los Angeles, CA (id = 35056)
ISP: HostCram LLC
Latency: 33.86 ms (0.12 ms jitter)
Download: 920.25 Mbps (data used: 1.4 GB)
Upload: 870.89 Mbps (data used: 1.0 GB)
Packet Loss: 0.0%
Result URL: https://www.speedtest.net/result/c/4fc8cb29-91f8-4d8b-bb1e-6463822d41aa
PHP Benchmark (2.66x faster SINGLE CORE performance, so 8x total performance. Average speed improvement for all of the tests is 2.3x)
HostCram
root@ubuntu:/usr/local/bin/phpbench# php bench.php -m=512 -t=300
-------------------------------------------------------------------------------------------
| PHP BENCHMARK SCRIPT |
-------------------------------------------------------------------------------------------
Start : 2021-09-26 21:39:40
Server : Linux/5.4.0-86-generic x86_64
Platform : Linux
System : Ubuntu 20.04.3 LTS
CPU :
model : 11th Gen Intel(R) Core(TM) i9-11900K @ 3.50GHz
cores : 3
available : 3
MHz : 3504MHz
Memory : 512 Mb available
Benchmark version : 1.0.39
PHP version : 7.4.22
available modules :
mbstring : yes
json : yes
pcre : yes
Max execution time : 300 sec
Crypt hash algo : MD5
-------------------------------------------------------------------------------------------
TEST NAME : SECONDS | OP/SEC | OP/SEC/MHz | MEMORY
-------------------------------------------------------------------------------------------
01_math : 0.996 sec | 1.00 MOp/s | 286.56 Ops/MHz | 4 Mb
02_string_concat : 0.135 sec | 56.89 MOp/s | 16.24 kOps/MHz | 128.84 Mb
03_1_string_number_concat : 1.011 sec | 4.94 MOp/s | 1.41 kOps/MHz | 4 Mb
03_2_string_number_format : 0.916 sec | 5.46 MOp/s | 1.56 kOps/MHz | 4 Mb
04_string_simple_functions : 0.829 sec | 1.57 MOp/s | 447.52 Ops/MHz | 4 Mb
05_string_multibyte : 0.596 sec | 218.00 kOp/s | 62.22 Ops/MHz | 4 Mb
06_string_manipulation : 1.541 sec | 843.65 kOp/s | 240.77 Ops/MHz | 4 Mb
07_regex : 1.062 sec | 1.22 MOp/s | 349.48 Ops/MHz | 4 Mb
08_1_hashing : 1.791 sec | 725.67 kOp/s | 207.10 Ops/MHz | 4 Mb
08_2_crypt : 5.378 sec | 1.86 kOp/s | 0.53 Ops/MHz | 4 Mb
09_json_encode : 1.285 sec | 1.01 MOp/s | 288.74 Ops/MHz | 4 Mb
10_json_decode : 2.013 sec | 645.82 kOp/s | 184.31 Ops/MHz | 4 Mb
11_serialize : 0.949 sec | 1.37 MOp/s | 390.91 Ops/MHz | 4 Mb
12_unserialize : 1.034 sec | 1.26 MOp/s | 358.96 Ops/MHz | 4 Mb
13_array_fill : 1.079 sec | 46.36 MOp/s | 13.23 kOps/MHz | 12 Mb
14_array_range : 0.296 sec | 338.21 kOp/s | 96.52 Ops/MHz | 12 Mb
14_array_unset : 0.914 sec | 54.72 MOp/s | 15.62 kOps/MHz | 12 Mb
15_loops : 0.374 sec | 535.41 MOp/s | 152.80 kOps/MHz | 4 Mb
16_loop_ifelse : 0.337 sec | 148.24 MOp/s | 42.31 kOps/MHz | 4 Mb
17_loop_ternary : 0.478 sec | 104.51 MOp/s | 29.83 kOps/MHz | 4 Mb
18_1_loop_defined_access : 0.174 sec | 114.82 MOp/s | 32.77 kOps/MHz | 4 Mb
18_2_loop_undefined_access : 1.500 sec | 13.33 MOp/s | 3.80 kOps/MHz | 4 Mb
19_type_functions : 0.403 sec | 7.45 MOp/s | 2.13 kOps/MHz | 4 Mb
20_type_conversion : 0.381 sec | 7.87 MOp/s | 2.25 kOps/MHz | 4 Mb
21_0_loop_exception_none : 0.016 sec | 252.97 MOp/s | 72.19 kOps/MHz | 4 Mb
21_1_loop_exception_try : 0.018 sec | 226.45 MOp/s | 64.63 kOps/MHz | 4 Mb
21_2_loop_exception_catch : 0.719 sec | 5.56 MOp/s | 1.59 kOps/MHz | 4 Mb
22_loop_null_op : 0.443 sec | 112.74 MOp/s | 32.18 kOps/MHz | 4 Mb
23_loop_spaceship_op : 0.382 sec | 130.75 MOp/s | 37.31 kOps/MHz | 4 Mb
26_1_class_public_properties : 0.042 sec | 118.36 MOp/s | 33.78 kOps/MHz | 4 Mb
26_2_class_getter_setter : 0.165 sec | 30.25 MOp/s | 8.63 kOps/MHz | 4 Mb
26_3_class_magic_methods : 0.374 sec | 13.36 MOp/s | 3.81 kOps/MHz | 4 Mb
-------------------------------------------------------------------------------------------
Total time: : 27.632 sec | 21.80 MOp/s | 6.22 kOps/MHz |
Current PHP memory usage: : 4 Mb
Peak PHP memory usage: : 125.54 Mb
Vultr
root@vultr:/usr/local/bin/phpbench# php bench.php -m=512 -t=300
-------------------------------------------------------------------------------------------
| PHP BENCHMARK SCRIPT |
-------------------------------------------------------------------------------------------
Start : 2021-09-26 21:39:08
Server : Linux/5.4.0-86-generic x86_64
Platform : Linux
System : Ubuntu 20.04.3 LTS
CPU :
model : Intel Core Processor (Skylake, IBRS)
cores : 1
available : 1
MHz : 3695.996MHz
Memory : 512 Mb available
Benchmark version : 1.0.39
PHP version : 7.4.22
available modules :
mbstring : yes
json : yes
pcre : yes
Max execution time : 300 sec
Crypt hash algo : MD5
-------------------------------------------------------------------------------------------
TEST NAME : SECONDS | OP/SEC | OP/SEC/MHz | MEMORY
-------------------------------------------------------------------------------------------
01_math : 1.817 sec | 550.23 kOp/s | 148.87 Ops/MHz | 4 Mb
02_string_concat : 0.292 sec | 26.34 MOp/s | 7.13 kOps/MHz | 128.84 Mb
03_1_string_number_concat : 1.693 sec | 2.95 MOp/s | 799.01 Ops/MHz | 4 Mb
03_2_string_number_format : 1.536 sec | 3.26 MOp/s | 880.92 Ops/MHz | 4 Mb
04_string_simple_functions : 1.600 sec | 812.38 kOp/s | 219.80 Ops/MHz | 4 Mb
05_string_multibyte : 1.327 sec | 97.93 kOp/s | 26.50 Ops/MHz | 4 Mb
06_string_manipulation : 3.522 sec | 369.13 kOp/s | 99.87 Ops/MHz | 4 Mb
07_regex : 2.493 sec | 521.42 kOp/s | 141.08 Ops/MHz | 4 Mb
08_1_hashing : 2.925 sec | 444.44 kOp/s | 120.25 Ops/MHz | 4 Mb
08_2_crypt : 23.519 sec | 425.18 Op/s | 0.12 Ops/MHz | 4 Mb
09_json_encode : 6.525 sec | 199.23 kOp/s | 53.90 Ops/MHz | 4 Mb
10_json_decode : 3.934 sec | 330.45 kOp/s | 89.41 Ops/MHz | 4 Mb
11_serialize : 1.984 sec | 655.18 kOp/s | 177.27 Ops/MHz | 4 Mb
12_unserialize : 2.117 sec | 613.98 kOp/s | 166.12 Ops/MHz | 4 Mb
13_array_fill : 2.416 sec | 20.69 MOp/s | 5.60 kOps/MHz | 12 Mb
14_array_range : 0.755 sec | 132.53 kOp/s | 35.86 Ops/MHz | 12 Mb
14_array_unset : 1.842 sec | 27.15 MOp/s | 7.35 kOps/MHz | 12 Mb
15_loops : 0.695 sec | 287.72 MOp/s | 77.85 kOps/MHz | 4 Mb
16_loop_ifelse : 0.657 sec | 76.05 MOp/s | 20.58 kOps/MHz | 4 Mb
17_loop_ternary : 1.584 sec | 31.57 MOp/s | 8.54 kOps/MHz | 4 Mb
18_1_loop_defined_access : 0.479 sec | 41.77 MOp/s | 11.30 kOps/MHz | 4 Mb
18_2_loop_undefined_access : 3.212 sec | 6.23 MOp/s | 1.68 kOps/MHz | 4 Mb
19_type_functions : 0.770 sec | 3.90 MOp/s | 1.05 kOps/MHz | 4 Mb
20_type_conversion : 0.764 sec | 3.93 MOp/s | 1.06 kOps/MHz | 4 Mb
21_0_loop_exception_none : 0.029 sec | 139.18 MOp/s | 37.66 kOps/MHz | 4 Mb
21_1_loop_exception_try : 0.032 sec | 124.75 MOp/s | 33.75 kOps/MHz | 4 Mb
21_2_loop_exception_catch : 1.311 sec | 3.05 MOp/s | 825.72 Ops/MHz | 4 Mb
22_loop_null_op : 1.293 sec | 38.67 MOp/s | 10.46 kOps/MHz | 4 Mb
23_loop_spaceship_op : 1.173 sec | 42.63 MOp/s | 11.53 kOps/MHz | 4 Mb
26_1_class_public_properties : 0.080 sec | 62.65 MOp/s | 16.95 kOps/MHz | 4 Mb
26_2_class_getter_setter : 0.316 sec | 15.82 MOp/s | 4.28 kOps/MHz | 4 Mb
26_3_class_magic_methods : 0.764 sec | 6.55 MOp/s | 1.77 kOps/MHz | 4 Mb
-------------------------------------------------------------------------------------------
Total time: : 73.458 sec | 8.20 MOp/s | 2.22 kOps/MHz |
Current PHP memory usage: : 4 Mb
Peak PHP memory usage: : 125.54 Mb
GTMetrix (on a very unoptimized, uncached BuddyBoss-based WordPress site)
| | HC | Vult | Diff |
|--------------------------|------|------|--------|
| GTmetrix Grade | B | B | -5% |
| Performance Score | 0.88 | 79% | -9% |
| Structure Score | 0.86 | 90% | 4% |
| Largest Contentful Paint | 1.7s | 2.3s | +645ms |
| Total Blocking time | 90ms | 55ms | -35ms |
| Cumulative Layout Shift | 0.12 | 0.11 | -0.01 |
I find the GT Metrix results to be wildly variable, but the LCP time aligns with a general 25-35% improvement in page load times when using the broswer dev tools network monitor.
All in all, I'm getting 5-8x faster performance (plus an extra 1GB of RAM) for nearly half the price. I would imagine if the website was stress tested with many connections, it could serve incomparably more connections than the Vultr one. If anyone knows how to do that easily, please let me know and I'll add the results.
Moreover, it seems to me that you can't find much better value anywhere. I'm generally preferential towards AMD and it seems like the Ryzen 5950x is a "better" processor, but I'm renting single cores, not a processor. And I'm not paying for the server's electricity. It also doesn't have ECC ram, but that shouldn't matter much for a non-mission-critical site like mine.
It is slightly annoying to have to wait for @Shakib to manually deploy the server and re-install the OS, but I don't foresee that ever needing to be done again.
I'm managing it through RunCloud for $8/month, which slightly irks me in comparison to the hosting fees. But DirectAdmin is too complicated, CyberPanel seems a bit finnicky, and the $1/month Pro offsite Backups (every 30 min) is by far the best solution that I can find.
Check out HostCram!
Comments
@nickc, Thanks for your kind words.
Additional free month has been added to your service!
I will also double your bandwidth for the month if at least 10 LET members comment "Double Bandwidth"
@Shakib Thanks! I don't need the bandwidth - feel free to do a random draw for the first 10 people who comment.
Okay.
Free VPS for 1 month!
If at least 10 LET members comment "Free VPS", i will also comment a coupon code that will work on our LXC-1GB plan > https://www.hostcram.com/vps
Whomever can claim it first, gonna have a Free VPS for 1 month!
Free VPS
Free VPS
Is no big news, that Vultr, DO and other cloud providers are overpriced.
Since I found LET, I never purchased a server from them again.
Since you can end up on different hardware at these providers, you may get better results if you are lucky, keep that in mind.
@Neoon Of course. It was just very striking to see how wide the gap is. Yet, it would appear from some other threads that the i9-11900 is more powerful per core than anything else, and the disk performance is pretty astounding as well. It would be hard to imagine a much more powerful setup, much less at these prices.
Keep in mind that the winner can get another free month (total 2 months) just by doing a review with benchmark on LET.
Free VPS
Free VPS
Free Vps
Free VPS
You should double ram if 10 people tag jbiloh and Dustin
Free VPS
Free VPS
Free VPS
one more benchmark. Single core comparison 2.5x higher. Not sure if you can then compare the multicore to the single core. Anyway, more confirmation of how much superior this is.
Unixbench
Hostcram
VultrHF
One more comment will unlock 🔑 🔓 the Free VPS coupon.
Edit: Keep an eye on this thread for the coupon code.
Free VPS
@panthera666 , @Warkos , @asaljeplak , @Andrews , @rafaelscs , @imgmoney , @giang , @gazmull , @Edmond , @ams
The Free VPS Coupon Code is: nickc
Will work with our LXC-1G plan > https://www.hostcram.com/vps
Whomever can claim it first, gonna have a Free VPS for 1 month!
Ugh, I created an account then realized I missed the place to put in the coupon code. Going back and the code already no longer works...
Who got the free VPS anyways?
same here, the code placed in unusual spot
Someone claimed. I asked him to confirm on LET so we all can know.
it was me
Too fast! And I accidentally ended up signing up for an account
Anyway to undo that by any chance? Sorry about that.
Delivered
I'd suggest you on keeping the account. You might get lucky someday very soon.
Another benchmark, using the
WP Performance Tester plugin. HostCram 2-3x faster. Probably just single core performance, given that it is PHP
HostCram
Vultr HF
Another test. Shows that Vultr is still quite good compared to most hosting out there (though it could be compared to an all-time database...). But HostCram is far superior.
Wordpres Hosting Benchmark Tool - CPU & Memory 2x, File System 2-4x, Network similar
HostCram
Vultr HF