Hosting Performance Contest – September 2018 Roundup (15 Hosts Tested: Fast and Wasted)

Share the knowledge...Share on Facebook
Facebook
Share on Google+
Google+
Tweet about this on Twitter
Twitter
Share on LinkedIn
Linkedin

hosting performance contest September 2018

Here’s the 32nd Monthly Hosting Performance Contest Roundup for September 2018!
I’ve been testing 15 hosts in a non-stop manner the whole month and I’ll show you the results below. In general, this contest’s results can be featured after the performance of one of the hosts as “Fast host does not mean the best performance”.

By the way, you can find the latest roundups here and the performance tables and charts here.

All monitored hosts are presented in the pricing categories according to the most affordable plan they have. Also, I bought the most affordable plans available at each hosting company to test hosting performance.

Let’s see how the hosts hosts performed in September 2018!


Common Information


As you may already know, I’ve been monitoring the performance of some of the best hosting providers I know as well as some other well-known hosts.

In short, since 2013 I’ve been picking out great hosts judging by real users reviews, professional hosting community opinion, long honorable business experience and other factors which make a great hosting.

And I buy anonymously hosting accounts with different hosts in order to closely monitor their performance (basically, uptime checked every minute and full page load time checked every 20 minutes) using a professional monitoring service monitis.com. I buy the most basic and the cheapest plan of each hosting. You can read about my monitoring methodology here.

And I’m happy to share these statistics with you so that you could see how these hosts perform and who is the best from the best!

You can find some real-time performance charts on my recommended hosts page for the hosting which I’ve been recommending and monitoring.

Also, you can see real-time monitoring charts for all hosts that I am monitoring on this page. In addition, the page contains monthly historical data on hosting performance (speed, uptime, satisfactory apdex) – very interesting and useful.

Besides, you may read more information about this Hosting Performance Contest on this page. I will be adding monthly and yearly Contest results on the page too.

And other monthly Hosting Performance Contest roundups are available from here.

By the way, here’s a disclosure: There are some affiliate links on this page. In other words, I get paid if you click on the links and make a purchase. All such links open in new window/tab; no software/program will be installed to your computer. (This is a standard notice required by hosting companies.)
Please note that although all hosts mentioned in this post are well-established and considered to be very good, I highly recommend not all of them. My recommended hosts are here.

 

September 2018 Hosting Performance Contest – Results

Here are the fifteen hosts that you can compare this month. I’ve broken down them in three groups: very affordable (below $5/mo), middle class (upto $8/mo) and more expensive ($8+/mo).

You’ll see a breakdown of hosting performance by prices in the sections below.

Let’s start from an overview of the hosting performance in the order of how the hosts performed in general regardless of how affordable or expensive the hosts are.

winner cup - hosting performance contest Roundup September 2018The winner of this month’s contest is GeekStorage!. By the way, it’s one of the best performing hosts historically among the shared hosts I present in this Contest.

In the Top-3 also go SiteGround and HawkHost.

What’s remarkable about the winner this month is that this host is not the most expensive. And all the winner hosts are in the different pricing categories.

Average full page load time of GeekStorage (the winner) was 1.2 seconds (it’s super fast). Its Uptime was almost perfect 99.99%. My site hosted with this host was available all the time during the whole month except 4.3 minutes. And the host’s Satisfactory Apdex was amazingly high 99.86% (i.e. 99.86% of the time during this month the full page load time was less than 2.5 seconds). Very impressive performance! All these parameters are much above the benchmarks specified for the greatest hosting performance. And since this hosting was the fastest host among the hosts which exceeded all the benchmarks determined for the greatest hosts GeekStorage wins the gold medal.

GeekStorage is one of comparatively affordable hosts in this contest, and it’s noticeable that it outperformed more expensive hosts! Very good job!

The silver medalist is SiteGround.

SiteGround performed in September 2018 as fast as 1.3 seconds which is impressively fast. It was slower than the leader by just 0.1 seconds. Its uptime was also almost perfect 99.99%. My site hosted with this host was available all the time during the whole month except 4.3 minutes. And the Satisfactory Apdex was also awesome – 99.81% (99.81% of all time the speed was faster than 2.5 seconds, and of course it’s much above the highest standards set for great hosting). Amazing performance!

The bronze medal goes to HawkHost.

In this month HawkHost took the 3d place in the Contest. My site was on average as fast as 1.37 seconds that is astonishingly fast. And HawkHost‘s uptime was very high 99.97%. My site hosted with this host was not available for just 13 minutes this month. Also, its Satisfactory Apdex was superb – 99.86% (i.e. 99.86% of all the time full page load time of my website hosted with this host was less than 2.5 seconds). Both metrics are much above the highest standards. Awesome!

VeeroTech took the 4th place. This host performed wonderfully well and satisfied both benchmarks set for great hosts. Its average speed was 1.48 seconds which is very fast. And its uptime (100%) was simply perfect and of course much above the benchmark (99.9%). My site hosted with this hosting was available all the time this month. The host’s Satisfactory Apdex (99.9%) was also very good and much above the highest standards (99%). Very solid performance!

StableHost this month took the 5th place. The hosting performed very well. It easily satisfied both benchmarks which are set for great hosting performance. The average speed was 1.56 seconds which is really fast. Its uptime (100%) was simply perfect and of course it satisfied the benchmark test (99.9%). My site hosted with this hosting was available all the time during this month. And the host’s Satisfactory Apdex (99.81%) was great and much exceeded the highest standards (99%).

GreenGeeks took the 6th place with 1.62 seconds speed, which is really fast. The host had almost perfect uptime (99.99%) which means that my site hosted with it was available for all the time this month except 4.3 minutes. This is much above the highest standards (99.9%). Besides, it had a perfect Satisfactory Apdex100%. It means that 100% of the time my testing site was faster than 2.5 seconds in this month. This is very well above the highest benchmark (99%) which is set for great hosts. Brilliant!

LunarPages goes to the 7th place. Its 1.83 seconds full page load time was fast. And its uptime (99.99%) was almost perfect and of course much above the benchmark level which is set for great hosts (99.9%). My site was not available for just 4.3 minutes this month. Besides, this host had a very good Satisfactory Apdex (99.76%) which was much above the benchmark (99%).

Other hosts could not meet at least one of the benchmarks set for the best hosting. However, the results of some hosts were very close to the benchmarks and pretty good in general.

In this month A2Hosting took the 8th place in the Contest. It was on average as fast as 1.37 seconds which is super fast. But since A2Hosting‘s uptime (99.88%) was a bit lower than the benchmark set for the great hosts (99.9%), this host does not go higher in the ranking. My site hosted with this host was not available during this month for 52 minutes. And its Satisfactory Apdex was brilliant – 99.95% (i.e. 99.95% of all the time full page load time of my website hosted with this host was less than 2.5 seconds). It’s much above the benchmark level set for the great hosts (99%).

Squidix has got the 9th place this month. It performed as fast as 1.44 seconds which is an awesome speed. But its two other metrics were not ideal. Its uptime (99.79%) was below the highest standards set for the greatest hosts (99.9%). My site hosted with this host was not available for 1.5 hours during this month. And the Satisfactory Apdex was below the highest standards as well – 97.96% (97.96% of the time this month the speed was faster than 2.5 seconds). It is a bit below the highest standards (99%).

InMotionHosting took the 10th place. Its average speed was generally great (1.87 seconds full page load time). But its uptime was 99.69%, which is below the benchmark level (99.9%). My site was not available for 2.2 hours this month. And the host’s Satisfactory Apdex (96.24%) was also below the highest standards (99%).

MochaHost took the 11th place.. Its speed was generally good (2.93 seconds). But its uptime (98.58%) was below the benchmark level (99.9%). My site was not available for 10.2 hours this month. And the host’s Satisfactory Apdex (92.44%) did not pass the benchmark level (99%) too.

Eleven2 got the 12th place. Its average speed (2.2 secs) was generally okay. Its uptime was 99.95%, which is superb and above the benchmark level (99.9%). My site was not available for just 21.6 minutes this month. The host’s Satisfactory Apdex (86.83%) could not get above the highest standards (99%) though.

HostWinds took the 13th place. The host had generally not bad speed (2.35 seconds on average). But it had not very good uptime (99.05%). It means my site hosted with this hosting was 6.82 hours offline. This is quite a lot below the uptime benchmark set for great hosts (99.9%). It’s Satisfactory Apdex (96.15%) was also below the benchmark (99%).

GlowHost went to the 14th place this month. This time it did not get the last place because another host screwed up more. GlowHost could not get higher because of its relatively slow speed. Its uptime (99.96%) was great and succeeded the benchmark (99.9%) which is set to determine the greatest hosts. My site was not available for 17.3 minutes this month. And the host’s speed was slow (3.17 seconds). And as expected, its Satisfactory Apdex (0%) was the lowest among other monitored hosts; and it was far from reaching the benchmark level (99%). The loading time of my test website was lower than 2.5 seconds during the whole month.

MDDHosting screwed up greatly with massive outage and therefore took the last 15th place. It’s ironical that this host could have taken the first place thanks to its fantastic speed performance. But the server maintenance failed badly due to a human factor and the host could not bring the users’ website back quickly. In addition there was a sudden problem with the snapshots of the users’ data. So, the restoring the users’ data took days instead of hours. After all, Michael Denney, the owner of the host did a great job by sincerely communicating to the clients and keeping us informed. So from the communicating point of view the host did a great job, but from a technical perspective the host screwed up for the first time like this in a decade.
And here are the performance characteristics for this host for September 2018:
Full page load time (1.05) seconds was super fast (actually the fastest among all the monitored hosts). And its Satisfactory Apdex (99.21%) was awesome and above the benchmark level (99%). But its uptime was severe (86.89%). My website was offline for almost 4 days.
By the way, the monitoring software reported downtime, but it did not report the actual offline state for my website. So I had to manually adjust (i.e. decrease) the uptime value compared to what Monitis reported.
I put this host on the last place this month because its downtime was unprecedented.

Generally, the competition of the monitored hosts this month was pretty tight as usual. And the incident with MDDHosting has brought in something unusual this month.

By the way, keep in mind, that for this Contest I pick out the hosts which are considered to be good, great and superb. These hosts are established businesses for many years and some of them are widely recommended in the Internet. So, the idea of this Contest is to determine the best performing hosts from a technical point of view, giving you objective information for making your own decision regarding hosts.

Let’s see now the results in tables and charts for more convenience, more information and more insight.

Here’s a table view to compare the Hosting Performance Contest results in September 2018:


Table column notes:
Place: The place a hosting has won in this Contest (the less the better).
Load Time: Average Full Page Load Time (the less the better). Checked every 20 minutes.
Uptime: Uptime (the more the better). Checked every minute.
Superb Uptime: If Uptime Benchmark (99.9%) passed (YES is very good).
Apdex-S: Satisfactory Apdex, i.e. how often a test website on a tested hosting was loading faster than 2.5 sec (the more the Apdex-S the better). Checked every 20 minutes.
Superb Apdex-S: If Satisfactory Apdex benchmark (99%) passed (YES is very good).
Color areas: Green is superb and above the highest standards. The greener, the better. Yellow is good, but below the highest standards. Orange is not very good; worse than yellow. Red is comparatively the worst.

Here’s a table with hosting prices:

 
1 year: Regular prices (after the first invoice) for 1-year plan.
2 years: Regular prices (after the first invoice) for 2-year plan.
3 years: Regular prices (after the first invoice) for 3-year plan.
Min Price: The minimum price officially available for the first invoice (can be for 1-year, 2-year or 3-year plan).
*StableHost discount code is given in my review.
Limited-time offers with more discounts are not included in these prices. Check out if there are special offers currently available by visiting the hosting websites.
Worthy: This column contains my recommended hosts. (Under review label means that the host performs well or/and has a good potential in winning prizes in the Contest, but I want to have more proved records of its technical and support performance.)

Here’s a table with places within pricing categories:

 
Since it’s not always correct to compare hosts from different price categories together, I’ve broken down the monitored hosts into three categories to compete within as you can see above in the table.

The prices presented in this table are regular prices (i.e. applied after the first invoice). Note that prices for the first invoice (1-,2- or 3-year plan) are usually less (see them in the previous table). And special promo prices are not included (check them out on the hosting websites).

And here are the charts with the Hosting Performance Contest results in September 2018:

1. The golden medalist: GeekStorage (it’s in my list of the recommended hosts; see my review).
2. The silver medalist: SiteGround (it’s in my list of the recommended hosts; see my review).
3. The bronze medalist: HawkHost (it’s in my list of the recommended hosts; see my review).
 

 

 

September 2018 Hosting Performance Contest Results – Notes


I use monitis.com services for uptime and full page load monitoring. The monitoring is performed from two locations in order to decrease chances of false alerts and make monitoring results more objective. You can read more about my methodology of monitoring here.

Here are the monthly monitoring reports screenshots:

Hosting Performance Contest September 2018 - uptime monitoring reportThese monitor checks are performed with 1-minute interval, which makes it great for detecting website uptime and downtime.

So, we need Uptime values from this table for finding the winner of our Contest.
By the way, downtime is detected if server response time (time-to-first-byte) is more than 10 seconds.

Other values (just for information):
Response time is not that important for this Contest, because I’ve got a better indicator monitored, which is Full page load time (see below).
Number of failures may be interesting to look at, but this is just an additional information to Uptime, which is more important for the overall hosting evaluation.

Six hosts on the screenshot have not passed the uptime benchmark test (99.9%) for the greatest hosting. The other hosts I’m monitoring were up more than 99.9% in September. Websites on the uptime-failed hosting were not available for more than 43.2 minutes in total this month.

And here’s below the screenshot of the continuous outage at MDDHosting for a few days. It was not reported by Monitis since technically my website was up (the server was up). But the website itself was replaced by the host with a temporary page.

MDDHosting outage

MDDHosting outage

Hosting Performance Contest September 2018 - full page load time monitoring report


These tests are performed with 20-minute interval from two locations. It makes it a perfect real-user performance monitor. This monitor’s activity is equal to 144 unique visitors per day (about 4,320 unique visitors per month).

Avg Test Duration is the Full page load time, which is very clear and important factor for the Hosting Performance Contest.

We don’t need Uptime from this table because we have got a better (more precise) Uptime indicator from the previous table.

Also, we don’t need Failures from this table, because it’s less precise than the same indicator from the previous table.

But we will need Apdex, particularly its S (Satisfactory) part. This shows how much time (in %) the website loaded faster than 2.5 seconds.

Conclusion

The best hosts which have demonstrated the best performance in September 2018 are the medalists: GeekStorage, SiteGround and HawkHost.

By the way, all the top winners have passed the strict benchmarks determined for great hosting!

As regards the winners in different pricing categories, here they are:
The best hosts in the higher pricing category ($8+/mo) in September 2018 go in the following order: SiteGround, GreenGeeks and A2Hosting.

The best three hosts in the middle pricing category ($5.01-8/mo) in September 2018 are GeekStorage, Veerotech, Squidix.

The best three hosts in the most affordable pricing category (up to $5/mo) in September 2018 are HawkHost, StableHost and LunarPages.

Please note, that I’m using the cheapest plans available on each of the monitored hosts.

It’s noticeable that some of the less expensive hosts outperformed hosts from more expensive categories.

And the featured (anti)hero of the Contest this month is MDDHosting. This host was the fastest this month but managed to failed very badly with uptime due to a massive continuous outage that happened after 10 years of pretty stable and great performance. Bad luck.


P.S.: My recommended hosts are here.
You can see real-time performance charts of the hosts I monitor as well as historical data on the hosting performance on this page.
Past and future Hosting Performance Contest results are (and will be) published on this page.
Other monthly roundups of this Hosting Performance Contest are available here.

Subscribe to Free Researches
Get smarter and work on your blog and small business more efficiently

subscribe
BTW, I respect your privacy, and of course I don't send spam, affiliate offers or trade your emails. What I send is information that I consider useful.

Share the knowledge...Share on Facebook
Facebook
Share on Google+
Google+
Tweet about this on Twitter
Twitter
Share on LinkedIn
Linkedin

Comments

  1. Thanks for all the work you do! Thank you.

    No word on Godaddy? The largest elephant in the room. Can’t find it anywhere in your writing…

    • Hey Jeff,
      I guess it’s too big to be noticed in the writings haha. It does not hide itself and it’s quite clear about this host from numerous clients reviews.
      Well, actually I do mention it somewhere in my article series devoted to hosting research.
      In very short, I don’t recommend this host. Although I knew personally a person (a business owner) who was quite happy with its services. He was wondering why everyone hates GD whereas he did not have noticeable issues with it. But it’s worth noticing that he had hired a technician to deal with the hosting 😉
      As regards its domain auctions it’s one of the biggest domain market and I used it personally multiple times. It did work well for me.

  2. “MDDHosting screwed up greatly” – yep. Entire data center destroyed by a typo. I didn’t even know that was possible. It was an ordeal for Michael and his team and their clients – including me. It is worth noting that MDD has been consistently awesome in the past, and because of this debacle has put in place measures to ensure consistent awesomeness going forward.

    • Yep, Kenny.
      The situation with MDDHosting reminds me the analysis of a malfunctioning in aviation:
      If there is just one mistake, then it ends up well since the system is ready for malfunctioning. But the more mistakes occur at the same time, the worse (or deadlier) it turns out.

      In case with MDD there was not just the problem with a sysadmin making a typo. There was also an issue with system snapshots. If the snapshots were functioning at the moment when the server admin screwed up, then everything would be running up again very quickly.
      But since snapshots were mistakenly turned off, MDD had to restore all backups which took days.

      Luckily the MDD backups were in order and there were no major data loss. The only loss is basically the time.

  3. Your research and continuous update on reliable cost effective hosting provider are really helpful. I am currently using stablehost for my few website and i am very satisfied with their services. Now i am planning to try another host Geekstorage. Can you please mention which hosting plan you used for testing of geekstorage performance and is there any measure difference between unlimited plan and performance plan ? Thanks for great write up

    • Hi Saurabh,
      Thanks a lot for your feedback.
      I’m using the cheapest plans on all hosts which I monitor. With GeekStorage I’m using Basic (limited) plan which is now called PX-1.
      There is a technical difference between unlimited and performance plans. In very simply words, performance plans offer better performance (RAM, IO speed), whereas unlimited plans allow storing more data.

It's important for me to know what you think

*