Strange performance compared to DO

Hello all. I've just setup a new 2gb ubuntu VPS that I'm looking to use to serve the API for my website. I've been using DO up till now but decided to try Linode out since the hardware seems to be better and prices are the same. I've setup a DO instance as well with the same specs in order to compare performance, and I'm running the exact same script on both instances. Both are fresh installs of Ubuntu 14.10.

My website is a block explorer for a cryptocurrency, so I've started out indexing the blocks from the client into my mongo database. The strange thing is that the Linode instance is noticeably slower! I'd say the DO instance is 15-25% faster right now just looking at the indexing progress.. I've checked the hardware specs, the Linode has 0.4 Ghz higher processor speed, and some basic SDD transfer tests give me about twice the transfer speed for the Linode instance.

Despite all that, it's slower!! One thing I've noticed is that the cpu use for the crypto wallet is pegged at 120% in my Linode instance but stays around 60% in DO.

Is there anything going on here that I've missed? Some settings or anything that could explain the performance gap?

1 Reply

If you're CPU bound, it's possible the other VPS has hardware extensions that your Linode doesn't. Here's what I get on Linode:

$ sed -n '1,/^$/p' /proc/cpuinfo
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 62
model name      : Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
stepping        : 4
microcode       : 0x416
cpu MHz         : 2800.036
cache size      : 25600 KB
physical id     : 0
siblings        : 1
core id         : 2
cpu cores       : 1
apicid          : 4
initial apicid  : 4
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu de tsc msr pae cx8 apic sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm constant_tsc rep_good nopl nonstop_tsc pni pclmulqdq ssse3 cx16 sse4_1 sse4_2 popcnt tsc_deadline_timer aes f16c rdrand hypervisor lahf_lm ida arat epb pln pts dtherm fsgsbase erms
bogomips        : 5602.40
clflush size    : 64
cache_alignment : 64
address sizes   : 46 bits physical, 48 bits virtual
power management:

The "flags" line indicates that, among other things, this CPU supports the AES and SSE4 instruction sets. Whether any of these are useful depends on the type of calculations the program is doing and if it's been compiled to use these instructions.

Linode uses Xen as its hypervisor; don't know what the other guys use. That could make a difference as well.

Reply

Please enter an answer
Tips:

You can mention users to notify them: @username

You can use Markdown to format your question. For more examples see the Markdown Cheatsheet.

> I’m a blockquote.

I’m a blockquote.

[I'm a link] (https://www.google.com)

I'm a link

**I am bold** I am bold

*I am italicized* I am italicized

Community Code of Conduct