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.
Comments
Hoooly shiit, NYC grabbing a 7GB file from Scaleway ... ~20MB/sec average - AWESOME!!
For comparison, I get around 6MB/sec on Dallas grabbing from Scaleway on VortexNode.
I can't knock Incero's network, great mix and routes are hella optimized.
Could someone post the rest of enabled cpu flags from cpu info?
Also,
openssl speed -evp aes-128-cbcI can not resist this. Plus, I love their network. And they're a cordial bunch as well!
It's your lucky day if AES is enabled :P
I got deja vu and remembered
https://www.lowendtalk.com/discussion/comment/1866797/#Comment_1866797
updating previous comment....
Finally, speedykvm almost beat time4vps without stupid I/O limit etc etc etc
Nice one!
Ok... Virmach $5/y:
dafuq!!! you know I can't passup on real storage deals...
geekbench: https://browser.geekbench.com/v4/cpu/2212220
# fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=1 --name=test --filename=test --bs=64k --iodepth=64 --size=1G --numjobs=2 --readwrite=randrw --rwmixread=75 test: (g=0): rw=randrw, bs=64K-64K/64K-64K/64K-64K, ioengine=libaio, iodepth=64 ... fio-2.16 Starting 2 processes Jobs: 2 (f=2): [m(2)] [100.0% done] [26048KB/9472KB/0KB /s] [407/148/0 iops] [eta 00m:00s] test: (groupid=0, jobs=1): err= 0: pid=1174: Wed Mar 29 07:44:09 2017 read : io=788032KB, bw=10178KB/s, iops=159, runt= 77425msec write: io=260544KB, bw=3365.2KB/s, iops=52, runt= 77425msec cpu : usr=0.20%, sys=0.61%, ctx=11950, majf=0, minf=7 IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.2%, >=64=99.6% submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.1%, >=64=0.0% issued : total=r=12313/w=4071/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0 latency : target=0, window=0, percentile=100.00%, depth=64 test: (groupid=0, jobs=1): err= 0: pid=1175: Wed Mar 29 07:44:09 2017 read : io=788480KB, bw=10169KB/s, iops=158, runt= 77541msec write: io=260096KB, bw=3354.4KB/s, iops=52, runt= 77541msec cpu : usr=0.23%, sys=0.59%, ctx=11960, majf=0, minf=6 IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.2%, >=64=99.6% submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.1%, >=64=0.0% issued : total=r=12320/w=4064/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0 latency : target=0, window=0, percentile=100.00%, depth=64 Run status group 0 (all jobs): READ: io=1539.6MB, aggrb=20331KB/s, minb=10168KB/s, maxb=10178KB/s, mint=77425msec, maxt=77541msec WRITE: io=520640KB, aggrb=6714KB/s, minb=3354KB/s, maxb=3365KB/s, mint=77425msec, maxt=77541msec Disk stats (read/write): vda: ios=24580/8160, merge=0/15, ticks=7127480/2532040, in_queue=9682068, util=100.00%@SpeedyKVM_Ryan
2 times my order has been canceled and refunded.
You should message us at [email protected] so that we can find the orders and identify the reason.
just finished the usual download bench:
very much satisfying so far ^^
Holy crap the benchmarks are just insane! I've never seen such good numbers on plans that cost twice as much.
From Toronto It's 11 hops to my VPS in NYC. For comparison it's 13 hops to a Luna node VPS hosted in Toronto!
Too bad you can only get 1 VPS with the code, I would have loved to move another production site there
There are 3 different codes. The one here, then one on Twitter and another on FB. Can use each of them once I believe.
Yah I was just testing that and it works. I'll submit a support ticket to see if it's okay. I don't want to piss them off by sneaking in on a technicality lol.
Better yet just ask Ryan here
https://www.lowendtalk.com/discussion/comment/2141025/#Comment_2141025
On the first page. Seems like they're endorsing it
Thanks @Harambe, I guess it's kosher!
Promo code valid in case of future upgrade from 1tb to 2tb?
Ouch! "Stupid" I/O limits are in place to guarantee you the best possible performance and price. Without said limitations either price will grow or service going straight to deadpool due to drive failures and massive service resource abuse by Plex / torrent owners.
Anyone would kindly explain pros and cons of raid60?
makes sense. But I wish the disk IO could be less strict on time4vps.
@SpeedyKVM_Ryan Can you spread multiple orders of the storage plan across multiple host nodes? Or is it too early to guarantee that at the moment?
About as good as Raid10 if your workload is mostly 95% reads. Not as good as raid10 for writes(small cpu computation penalty for parity writes)
raid60 is raid6 with striping, so writes would be slightly better than plain raid6.
With disks larger than 2TB, Raid6 is the way to go. (with raid60 being the faster upgrade)
Ultimately everything depends on the kind of io loads on customer vms. their available IOPS looks similar to time4vps arrays.
Seems perfect for a static site project I'm thinking. Thanks!
@SpeedyKVM_Ryan automatically cancelled order and refunded. No one reply ticket or email.
speedykvm don't put such limitation, but still able offer better deal. on your storage vps, to check folder size take more than few minute (test it your self
du -sh). I run some custom script to validate my backup, which take ages to finish.For cold storage or small size backup (under 1 GB per folder) that still usable. I'd love to pay more if there's option to waive I/O limitation.
btw good luck with your storage vps, maybe that not fit for my need.
@time4vps 's product is great for casual Linux iso circulation and I'm sure that speedykvm doesn't intend to attract that crowd on its product, hence speedykvm can provide a better IO with a stricter policy for the rest of the client niches.
Win-win for everyone involved!