WebFaction
Community site: login faq
0
1

Recently WebFaction migrated my account on shared server from Softlayer Dallas to HEG/server4you datacenter in St.Louis. And I had an option to reverse this migration too.

I compared the new server at HEG and its really impressive in terms of speed. I got I/O about 400 MB/s which was as good as Softlayer and the CPU are also good 40 cores (It's seems 20 cores with hyper threading) and massive 126 GB RAM. What alerts me is the network connectivity as HEG data center is using Telia as the only 1 major bandwidth provider apart from Cogent as secondary. Whereas Softlayer uses Level3, NTT, At&t, Telia, Tata as its direct peers!

How its good connecting Europe, Asia and Japan/Australia with Telia network at HEG compared to Softlayer?

How are you guys doing on your new server at HEG? Any tests done? Or are you happy staying with Softlayer servers?

asked 19 Dec '16, 05:59

alexmedia's gravatar image

alexmedia
111
accept rate: 0%


We've actually been using servers at this datacenter for some time, putting all new clients on these machines and tracking performance for months. We've been pleased with the results and it was only afterward that we decided to migrate existing servers to the new datacenter. This is mainly for the significantly improved hardware. Having a lot more RAM and CPU power has helped improve the hosting experience for users.

As for network connectivity, I think it's always going to be the case that a superset of five network pairings is superior to a subset of one. That said, we haven't been receiving a lot of feedback about reduced performance after a server migration.

To get a feel for the difference in global response times, you can use a multi-ping tool like this one. Comparing a random softlayer USA server (75.126.137.89) with an HEG one (207.38.86.9) showed (in my test) that the average RTT from their pool of servers decreased by 1.9ms and std deviation decreased by 2.5% from 77.5ms to 75.5ms -- those are both buried in the statistical noise floor, and it follows that for most users the differences are negligible.

Where there's going to be a difference is when you have a website visitor or client whose own ISP has fairly poor network pairings and they also don't pair with Telia. For those users, one of the other parings at Softlayer may help -- a lot. It's difficult to collect data on exactly how many users and their ISPs don't pair well with Telia, but given that we haven't seen many complaints over a few months, it doesn't seem to be particularly prevalent at all.

So, in summary, what you're likely to see is a very small subset of users experience much worse network performance while everyone else is unaffected, meanwhile large disruptions to service which affect all visitors resulting from server load should be reduced due to much better server specifications. On balance it's a good move, and we hope you find the new server environment amicable.

permanent link

answered 24 Dec '16, 02:01

ryans's gravatar image

ryans ♦♦
5.0k42652
accept rate: 43%

Your answer
toggle preview

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here

By RSS:

Answers

Answers and Comments

Markdown Basics

  • *italic* or _italic_
  • **bold** or __bold__
  • link:[text](http://url.com/ "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported

Question tags:

×108

question asked: 19 Dec '16, 05:59

question was seen: 393 times

last updated: 24 Dec '16, 02:01

WEBFACTION
REACH US
SUPPORT
AFFILIATE PROGRAM
LEGAL
© COPYRIGHT 2003-2016 SWARMA LIMITED - WEBFACTION IS A SERVICE OF SWARMA LIMITED
REGISTERED IN ENGLAND AND WALES 5729350 - VAT REGISTRATION NUMBER 877397162
5TH FLOOR, THE OLD VINYL FACTORY, HAYES, UB3 1HA, UNITED KINGDOM