Moz Q&A is closed.
After more than 13 years, and tens of thousands of questions, Moz Q&A closed on 12th December 2024. Whilst we’re not completely removing the content - many posts will still be possible to view - we have locked both new posts and new replies. More details here.
Using a third party server to host site elements
-
Hi guys -
I have a client who are recently experiencing a great deal of more traffic to their site. As a result, their web development agency have given them a server upgrade to cope with the new demand.
One thing they have also done is put all website scripts, CSS files, images, downloadable content (such as PDFs) - onto a 3rd party server (Amazon S3). Apparently this was done so that my clients server just handles the page requests now - and all other elements are then grabbed from the Amazon s3 server. So basically, this means any HTML content and web pages are still hosted through my clients domain - but all other content is accessible through an Amazon s3 server URL.
I'm wondering what SEO implications this will have for my clients domain? While all pages and HTML content is still accessible thorugh their domain name, each page is of course now making many server calls to the Amazon s3 server through external URLs (s3.amazonaws.com).
I imagine this will mean any elements sitting on the Amazon S3 server can no longer contribute value to the clients SEO profile - because that actual content is not physically part of their domain anymore. However what I am more concerned about is whether all of these external server calls are going to have a negative effect on the web pages value overall. Should I be advising my client to ensure all site elements are hosted on their own server, and therefore all elements are accessible through their domain?
Hope this makes sense (I'm not the best at explaining things!)
-
Hello Zeal Digital,
I use a CDN (Content Delivery Network) for images, CSS and javascript.
Doing that adds only about $10 to cost per month for a site that had around 800,000 pageviews per month.
You have complete control over the images. If there is a problem, you can force the CDN to flush a file and reload it from the source. You add code to your .htaccess file that tells the CDN how long to store images before fluching them and getting a new copy. It is all automated, there is generally no work for you to do. I host with softlayer.com and this is part of their service.
The change from self-sourced images, css and scripts had a massive improvement on the server.
- it is a 16-processor linux box with twin 15,000rpm SCSI drives and 12Gb RAM - it is quite fast!
Page delivery times improved by 1-2 seconds.
The server now is so lightly loaded that it could be downgraded to save more money.
It has zero effect on SEO. The CDN is accessed using a CNAME.
- static.domain.com - so don't worry about it looking like components are from other places.
The CDN has servers all over the world, so no matter where the visitors are, it is only a few hops for them to get most of the content, making it much faster for someone in Australia who would normally pull images from a server in the USA.
Your only problem with Amazon S3 is that they have crashed it a few times, but other than that, it is a good thing to do.
I wouldn't advise them to self-host, unless you want to increase their costs, server loading and page delivery times.
-
Great advice, cheers Jeffery!
-
I work with a number of high traffic sites (TB's of data each day, 10's millions page views/month). With many of these sites, we have offloaded static content to either dedicated static content servers (typically cloud based so we can scale up and down) or to content deliver networks. I've not had anyone report any SEO impact.
In contrast, they often see user engagement (page views/user), repeat visitors, and other traffic metrics improve. Users like fast sites. Also, Google apparently likes fast sites too, so while I've not seen it, you could actually get a boost in your SERPs due to faster loading pages.
If you break down a modern web page, you will find numerous elements required. Dozens of images, CSS, javascript as well as the page itself. All of these items require a request to the web server.
With some graphic intensive sites, I've seen as much as 95% of all web server requests (HTTP requests) be attributable to static content. By moving these HTTP requests to other systems, you free your primary server to handle the application. This provides a better user experience and improves scalability.
Content Delivery Networks
I do not use Amazon's Web Services so I do not know specifically what they offer. But here are two CDN's Ihave used with good success:
Internap:
http://www.internap.com/cdn-services-content-delivery-network/
Edgecast:
One method I look for is called "origin pull." With this method, you do not have to upload files to the CDN. The CDN will fetch them automatically from your site as needed. I found this is much easier to manage on sites that have frequent content updates.
-
Hosting images externally never had any impact on cases I had a chance to observe. The only problem I can think of is that you lose control over loading times or if somebody takes an image and links (credits) the image hosting domain instead of your domain.
-
Couple of notes for you
- There isn't any SEO impact on WHERE the data is loaded from. Look at any major website (especially one that ranks well) and they're openly using content delivery (like Akamai, Amazon S3/Cloudfront, etc) for static content. This is good business practice because it takes that load off your web server and often places the content closer to where the client is. Faster content delivery can help SEO if you have a slow server.
- If they're using the raw S3 buckets I would HIGHLY suggest signing up for Cloudfront. There's two benefits to doing this. First, you put the content into Amazon's cloud, where it is more readily available. Second, you can use domain aliasing to help obscure the source. For instance, let's say you have an images bucket. You could add a CNAME DNS record for images.yourdomain.com and then put that into your source code. You can still see where the DNS takes you, but it's not obvious to the general public. The cost difference between raw S3 delivery and Cloudfront is negligible.
Oh, and I use Amazon Cloudfront for my delivery. Never had any SEO issues with doing so.
-
I don't recomend to have the resources and database to other server than files, it makes some flood traffic between servers, the resources are harder to load and the site optimum speed is decreased. Also you can't compress this content so they are downloaded independently.
Got a burning SEO question?
Subscribe to Moz Pro to gain full access to Q&A, answer questions, and ask your own.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
Using GeoDNS across 3 server locations
Hi, I have multiple servers across UK and USA. I have a web site that serves both areas and was looking at cloning my sites and using GeoDNS to route visitors to the closest server to improve speed and experience So UK visitors would connect to UK dedicated server, North America - New York server and so on Is this a good way or would this effect SEO negatively. Cheers Keith
Technical SEO | | Keith-0071 -
Server Connection Error when using Google Speed Test Insight and GTMetrix
Hi Guys, Recently got into the issue when testing load speed of my website (https://solvid.co.uk). Occasionally, Google Speed Insights gives me a server connection error which states _"PageSpeed was unable to connect to the server. Ensure that you are using the correct protocol (_http vs https), the page loads in a browser, and is accessible on the public internet." Also, GTMetrix gives me an error as well, which states the following: "An error occurred fetching the page: HTTPS error: SSl connect attempt failed" All of my redirects seem to be set-up correctly as well as the SSL certificate. I've contacted my hosting provider (godaddy), they are saying that everything is fine with the server and the installation. Also, tried in different browsers in incognito mode, still gives me the same error. Until yesterday I haven't had such a problem. I've also attached the error screenshot links. I would really appreciate your help! Dmytro UxchPYR M52iPDf
Technical SEO | | solvid1 -
Mobile site ranking instead of/as well as desktop site in desktop SERPS
I have just noticed that the mobile version of my site is sometimes ranking in the desktop serps either instead of as well as the desktop site. It is not something that I have noticed in the past as it doesn't happen with the keywords that I track, which are highly competitive. It is happening for results that include our brand name, e.g '[brand name][search term]'. The mobile site is served with mobile optimised content from another URL. e.g wwww.domain.com/productpage redirects to m.domain.com/productpage for mobile. Sometimes I am only seen the mobile URL in the desktop SERPS, other times I am seeing both the desktop and mobile URL for the same product. My understanding is that the mobile URL should not be ranking at all in desktop SERPS, could we be being penalised for either bad redirects or duplicate content? Any ideas as to how I could further diagnose and solve the problem if you do believe that it could be harming rankings?
Technical SEO | | pugh0 -
Host sitemaps on S3?
Hey guys, I run a dynamic web service and I will start building static sitemaps for it pretty soon. The fact that my app lives in a multitude of servers doesn't make it easy to distribute frequently updated static files throughout the servers. My idea was to host the files in AWS S3 and point my robots.txt sitemap directive there. I'll use a sitemap index so, every other sitemap will be hosted on S3 as well. I could dynamically mirror the content from the files in S3 through my app, but that would be a little more resource intensive than just serving the static files from a common place. Any ideas? Thanks!
Technical SEO | | tanlup0 -
Links from the same server has value or not
Hi Guys, Sometime ago one of the SEO experts said to me if I get links from the same IP address, Google doesn't count them as with much value. For an example, I am a web devleoper and I host all my clients websites on one server and link them back to me. Im wondering whether those links have any value when it comes to seo or should I consider getting different hosting providers? Regards Uds
Technical SEO | | Uds0 -
How to extract URLs from a site (without bringing the server down!)
Hi everybody. One of my clients is migrating to a new ecommerce platform, and we need to get a list of urls from the existing site to start mapping out the 301 redirects. Usually, I'd use a tool like Xenu or Integrity to crawl and output a list. However, the database and server setup is so bad that it can't handle the requests from these tools and it sends the site down. This, unsurprisingly, is one of the reasons for the migration. Does anybody know of a way to get a full list of urls without having to make a bunch of http requests which will kill the site? Any advice would be much appreciated!
Technical SEO | | neooptic0 -
Does google use the wayback machine to determine the age of a site?
I have a site that I had removed from the wayback machine because I didn't want old versions to show. However I noticed that in many seo tools the site now always shows a domain age of zero instead of 6 years ago when I registered it. My question is what do the actual search engines use to determine age when they factor it into the ranking algorithm? By having it removed from the wayback machine, does that make the search engines think the site is brand new? Thanks
Technical SEO | | FastLearner0