Web hosting providers have become a huge community, and there is very little that distinguishes one hosting company from the other. The features and core plans offered are basically the same, and pricing is no longer the deciding factor. In fact selecting a host based on lower price can turn out to be more expensive in the longer run, because the reliability issues and loss of sales may be a possibility when the servers are down and the website is not available.
Importance of Reliability-Uptime
It is a daunting task to finalize a host for your site from the thousands of web hosting providers available on internet. As a result you have to use a trial and error method before settling for a reliable one. Although there are many things to consider for deciding the web host, one of the main commodity to look for is uptime or reliability. Redundancy is always at the heart of any web hosting company’s reliability. This ensures that if there is a problem there always is an alternative that ensures continuous working of the website as seamlessly as possible. Most hosting providers have redundant network connections. High speed pipes are used for routing data from the server to your browser. Redundant multiple servers are very rare and expensive, as they need routing equipment which is costly and previously used in critical applications of some Fortune 500 companies.
Why Use Multiple Servers?
The DNS (Domain Name Server) ’round robin’ feature allows selection of a specific IP address from a pool of IP addresses, when a DNS request is sent. Remember, DNS database maps a host name to its IP address. So, rather than remembering a series of IP addresses, you can just type www.yourdomain.com in the browser to reach your site. Normally it takes 2-3 days to spread the word about your DNS on internet. Due to this when a domain name is registered or transferred, it is not immediately available to a person browsing on the net. This delay has a negative effect on the security benefits of hosting your site on multiple servers; as your website will be down for a couple of days in case there is some problem with one server. You will need to change your DNS for the second server and wait for days together before the routers pick up the change. This problem is resolved with the ’round robin” DNS approach which will map your domain name with multiple IP addresses.
How it Works
Now you can select the web hosting provider and employ DNS round robin technique along with ‘failover monitoring’. This process begins with the hosting provider setting up your website on multiple servers with dissimilar IP blocks allocated to them. So, the domain name will have multiple IP addresses. The failure monitor now dispatches the data to the URL you specify, and by looking for this text in results. If one of the IP addresses is returning an error, the system pulls that IP address out from the list. After this the DNS points the domain name to the working IP addresses. If this IP comes back online, it is re-enlisted in the IP pool. Effectively this means your website is always online even when the server is down.
Failure Detection/Recovery Time
The failure detection along with recovery time in this case is about fifteen minutes. This time depends on your site’s speed, nature of the problem and how long the ISPs can save your DNS info. The failure monitor can manipulate this time by lowering TTL cache settings. Other ISPs will also use the same settings for deciding the time to cache this DNS info. How frequently the data is synchronized between the site’s servers is the web hosting provider’s responsibility. All this becomes more complicated with the involvement of user sessions and DBs.
Hardware Based Failure Monitoring System
Another solution is the expensive hardware based failure monitoring system. It points a virtual IP address to other ISPs by selecting from a number of standalone IP addresses on various web servers. This method eliminates the need of ISPs caching your info.
Importance of Reliability-Uptime
It is a daunting task to finalize a host for your site from the thousands of web hosting providers available on internet. As a result you have to use a trial and error method before settling for a reliable one. Although there are many things to consider for deciding the web host, one of the main commodity to look for is uptime or reliability. Redundancy is always at the heart of any web hosting company’s reliability. This ensures that if there is a problem there always is an alternative that ensures continuous working of the website as seamlessly as possible. Most hosting providers have redundant network connections. High speed pipes are used for routing data from the server to your browser. Redundant multiple servers are very rare and expensive, as they need routing equipment which is costly and previously used in critical applications of some Fortune 500 companies.
Why Use Multiple Servers?
The DNS (Domain Name Server) ’round robin’ feature allows selection of a specific IP address from a pool of IP addresses, when a DNS request is sent. Remember, DNS database maps a host name to its IP address. So, rather than remembering a series of IP addresses, you can just type www.yourdomain.com in the browser to reach your site. Normally it takes 2-3 days to spread the word about your DNS on internet. Due to this when a domain name is registered or transferred, it is not immediately available to a person browsing on the net. This delay has a negative effect on the security benefits of hosting your site on multiple servers; as your website will be down for a couple of days in case there is some problem with one server. You will need to change your DNS for the second server and wait for days together before the routers pick up the change. This problem is resolved with the ’round robin” DNS approach which will map your domain name with multiple IP addresses.
How it Works
Now you can select the web hosting provider and employ DNS round robin technique along with ‘failover monitoring’. This process begins with the hosting provider setting up your website on multiple servers with dissimilar IP blocks allocated to them. So, the domain name will have multiple IP addresses. The failure monitor now dispatches the data to the URL you specify, and by looking for this text in results. If one of the IP addresses is returning an error, the system pulls that IP address out from the list. After this the DNS points the domain name to the working IP addresses. If this IP comes back online, it is re-enlisted in the IP pool. Effectively this means your website is always online even when the server is down.
Failure Detection/Recovery Time
The failure detection along with recovery time in this case is about fifteen minutes. This time depends on your site’s speed, nature of the problem and how long the ISPs can save your DNS info. The failure monitor can manipulate this time by lowering TTL cache settings. Other ISPs will also use the same settings for deciding the time to cache this DNS info. How frequently the data is synchronized between the site’s servers is the web hosting provider’s responsibility. All this becomes more complicated with the involvement of user sessions and DBs.
Hardware Based Failure Monitoring System
Another solution is the expensive hardware based failure monitoring system. It points a virtual IP address to other ISPs by selecting from a number of standalone IP addresses on various web servers. This method eliminates the need of ISPs caching your info.
Comments
Post a Comment