If you're considering whether you can afford running more than one server for your business, you're likely to discover that you can't afford not to. Even with the additional operational expenses, you’ll be better off having redundant servers because redundancy minimizes downtime for your critical applications and reduces the costs incurred by outages, which can be huge. The price of downtime is always too high for any business.
Like it or not, IT downtime is a fact of life. Systems fail. Outages happen. The failure rate of cloud servers is roughly 2% annually, for example. The larger the installation, the more frequently outages will occur. The best way to deal with these inevitable worst-case scenarios is to be prepared and minimize the downtime as much as possible.
Counting the cost of downtime
According to market research firm IHS, information and communication technology (ICT) downtime costs North American businesses $700 billion per year, resulting from outages or performance degradations on servers, applications and networks. IHS calculates that downtime costs $1 million per year for an average medium-sized business and more than $60 million for a large company, while the biggest costs are lost productivity and lost revenue.
There are myriad factors that affect the calculation of downtime costs, such as size and type of your company, the nature and duration of the outage or degradation as well as the amount of built-in redundancy and any other mitigation or disaster recovery tools used.
No matter how big or small the system failure is, it will cost your business in a variety of ways. Some costs are tangible while others are intangible and more difficult to put a value on.
The most obvious cost to your business is lost sales due to a web server failure or performance degradation. If customers can’t access your site because it’s down, then they can’t buy your products. If it takes too long to load, customers will give up trying, abandon your site and go to your competitors. Studies have shown that 40% of people will abandon websites that take more than 3 seconds to load.
Amazon has found that every 100 milliseconds of latency costs then 1% of their sales. For the world’s largest online retailer, lost sales add up quickly. More generally, if an e-commerce site makes $100,000 per day, a 1-second page delay will cost $2.5 million in lost sales each year.
Other costs are related to the toll IT downtime takes on your employees. When critical business applications like email or remote desktop service fail, your employees are prevented from working. Productivity can quickly drop to virtually zero as your employees are not able to perform tasks that will generate revenue for your business.
When your website fails or is too slow, that creates a negative perception among potential and existing customers that there could be something wrong not just with the site but with your company. Poor online performance quickly erodes trust among users and damages your company’s reputation. Performance also affects your search engine optimization (SEO) scores, which determines how easily people can find your business online. Low scores can result in fewer site visits and lower conversions over time.
Redundancy to the rescue
It’s not enough to simply duplicate server instances to create system redundancy. Effective redundancy leverages load balancing to distribute workloads across servers and keep your critical services up and running in any event.
Load balancing enables your business to:
- Avoid downtime from scheduled maintenance by rerouting traffic around servers that are temporarily out of service. This ensures employees and customers are not affected by the maintenance.
- Handle traffic surges from planned and unplanned events by distributing traffic so that servers don’t become overloaded and continue to run at peak performance.
- Manage failures by detecting server problems early and routing traffic to healthy servers that are still operating.
Different types of redundancy
There are different ways to achieve high-availability and keep your critical applications always on. Two examples of redundancy modes are Active-Active and Active-Passive.
In Active-Active Redundancy mode, two or more instances of the same network device or appliance serve clients simultaneously and interchangeably. They are configured to run in exactly the same way so that if any instance goes offline, another instance seamlessly starts serving requests without interrupting the client sessions.
In Active-Passive Redundancy mode, two or more identical instances of the same network device or appliance operate simultaneously but only the active instances serve client requests. If an active instance goes offline, passive instances kick into operation as active nodes to service requests.
IT uptime is imperative for modern businesses today. The fallout from server failures includes tangible costs from lost revenue and productivity as well as intangible costs that can cause lasting reputational damage to your business. Redundancy coupled with load balancing minimizes the high costs of IT downtime.
To learn how Snapt’s software load balancer as part of the Application Delivery Controller (ADC) avoids the cost of downtime and keeps your applications up and running, start your free trial today.