Congratulations! You’ve gone viral! Your content pierced the public psyche and now everyone is coming to your site for the same thing all at the same time.
It was first coined as the “Reddit Effect,” when thousands of people would read something on Reddit and share it on social media causing a frenzy of interest. Minutes later, your website is crashing and the only thing everyone beyond the first few hundred people get to see is an error message or, worse yet, nothing at all. This unfortunate occurrence was aptly coined the “Reddit Hug Of Death.”
After people have experienced your viral content is when the most engaged (and most valuable) visitors discover you and your brand. Additionally, other content on your site suddenly gains steam.
If properly prepared, the Reddit Effect has pulled brands and individuals out of obscurity and launched them."
For that to happen, you need to know how to prepare for spikes in traffic.
What Happens During a Reddit-Hug-Of-Death?
If you haven’t got the mechanisms in place with your cloud provider to handle sudden spikes in traffic, the result is a complete shutdown of the site. Implementing these mechanisms at the moment of a sudden spike can take time, can drastically increase costs, and can cause you to completely miss the opportunity.
What if it happens overnight? You could have missed the moment altogether.
From a cloud perspective, the capacity for visitor access to the content is based on the plan you’ve chosen, generally conforming to the normal traffic to the site in any given period. Data is backed up so it isn’t lost, but if traffic suddenly spikes, a crash is inevitable.
Downtime can kill momentum and unless there are adjustments made quickly, the momentum will be gone. This is not the time you want to test the level of service of your cloud provider. It is best to recognize some of the things that could trigger the Reddit Effect and choose a provider who offers services that can reasonably avoid costly downtime.
How to Avoid the Reddit Hug of Death
Sometimes it’s impossible to plan for sudden a spike in traffic. As we all know, the internet can be the ever-expanding ouroboros snake eternally feeding upon itself. However, your activities can inform your choices of service plans and mechanisms you implement with your cloud provider.
- Creating Content: If you create blogs, video, or social media content intended to go viral, experts recommend having a high availability environment with a server cluster to avoid downtime due to high traffic.
- Newsworthy Activity and Marketing: When engaging in national media appearances, social media ad buys, and creating digital marketing traffic, a high availability environment and fault tolerance using Managed Dedicated Servers will help ensure uptime during fluctuations in traffic.
- Unwanted Virality and PR: When things go bad, the need to control the message is important and a site crash destroys that chance. Crisis managers cannot rely on news or outside sources to correctly disseminate your full message, so high availability is key.
Your data is housed on a server and visitors go through a lot of levels at the speed of light to get there. If you employ load balancing technology, once they reach your cloud provider, they are first connected to either managed shared or dedicated load balancers. Using one of many available traffic optimization algorithms, these direct each distinct visitor to one of the web servers in your infrastructure.
How does load balancing offset the Reddit Effect? As traffic to your content swells, new servers can be quickly and easily added to the pool of load-balanced servers."
Having this infrastructure in place up front can save a lot of time and lost revenue.
Managed Shared Load Balancers
Managed shared load balancers are intended to mitigate growth beyond a single web server. This mechanism is highly reliable, customizable, and you’ll receive full redundancy with automatic failover for your web application.
Managed Dedicated Load Balancers
Managed dedicated load balancers offer the benefits of shared load balancing with additions such as advanced traffic scripting options, a complete API, and high-performance SSL. It ensures that a full set of resources are committed to your infrastructure at all times and scale at your own pace. This increases performance and protects against latency or bottlenecks with dedicated hardware that can cause downtime.
High Availability and Fault Tolerance
A high availability environment is achieved by allocating more than one server and a floating IP address with data replication to keep them in sync as a failover mechanism. Redundancy is the key to allowing traffic to run freely without downtime, even if a server goes down.
No network is perfect, and an outage is possible for reasons far beyond the Reddit Effect, so having fault tolerance in place is an excellent failsafe. Using a Storage Area Network (SAN), fault tolerance houses the basic core functionality of the site on a server, enabling visitors to access the system in the event of the failure of one or more components.
Your Popularity Shouldn’t Crush You
If you’re creating content, advertising, or increasing your social media activity, then definitely consider reviewing your current cloud services package. Along with increased traffic comes increased exposure and scrutiny, so extra security should also be a consideration.
Putting all of the mechanisms in place in anticipation of your burgeoning popularity isn’t aspirational; it’s practical. Keep in mind load balancing and fault tolerance in a high availability environment when creating a cloud services package.
Is your website’s availability crucial to your business? Get a highly reliable host like Liquid Web to keep your site scalable, secure, and fast loading. Browse our hosting plans to get started today.
A self-professed pirate captain with two decades of leadership experience, Jerry has led teams from 60+ cooks and chefs to 16 networking engineers. He brings those years of experience to his current role as Product Manager at Liquid Web, focusing on networking and security products. When not working or sleeping, Jerry can usually be found eating and having a good conversation with good people.
Keep up to date with the latest Hosting news.