In thinking through some blog topics recently amidst some 2012 planning, the word ‘expectations’ came to mind. What do we expect to be next year? What do we expect of ourselves and most importantly, what do the clients and customers of Dyn expect out of us?
We are an Internet Infrastructure company that focuses on website uptime and best-in-class email delivery for those that want to meet the expectations of the masses. When people visit Netflix to organize their queue, they expect to be able to do that without pain. When a Tumblr user wants to blog about their kids, the expectation is the CMS Is available to do that. When Mill33 sells email services to clients, their expectation is that clients will be able to send large volumes of email without running into problems.
As we have continued to evolve on the Internet, expectations have been raised across the board. This is how we try and suggest that they are met.
The Internet is becoming more resilient and faster: more connectivity, more fiber in the ground and better technology to manage it all is making content move faster. As the reliability of the Internet gets better, user expectations get bigger. No longer do we want sites to eventually load — we want them to load fast.
Those with slow sites on desktops or mobile devices (now about 5-10% of the web browsing population) will see their audience just move on to the competition and possibly never return. Speed affects search rankings, which is another big reason these types of things matter.
Our managed DNS services help eliminate issues and speed content up. We help deliver that first byte sooner since DNS lookups take place before HTML and images are loaded. As an IaaS provider, we have to be both up and fast all the time. Other people use our services to stay nimble and be able to move things around. We’ve been fortunate enough that many of them have shared their stories with us on how we’ve been able to help them.
In today’s Internet, there’s no excuse for website sluggishness or downtime. Your readers, customers and partners expect more — regardless of what type of business you are.
Email Delivery Infrastructure
Dyn is a few years into providing rock solid email delivery/SMTP services and earlier this year, we launched DynECT Email Delivery, a platform that delivers legitimate email to inboxes for enterprise-level email senders and gives great insight into what happens along the way.
Email is still a community protocol; anyone can send an email to anyone else. When you are a small volume sender, your messages usually aren’t blocked or are otherwise discarded because no one is watching. As your sending volume grows, ISPs and the gateway to inboxes start to take notice. They watch rates of sending, content, reputation and many other factors. Frequently, it’s a wholesale block.
We discovered that we need to touch a lot of the process around email to help our customers be more effective senders. There is the technical construction of email, the configuration of infrastructure around it and the content/goals that senders have. Like DNS, our customers’ infrastructure has many touchpoints with DNS and we gain a unique vantage point to help people speed up their site or application.
Like the early days before fairly global Internet reachability, there was nuance in the delivery of packets on the Internet. Today, that still plays out with peering and settlement free transit but is largely unnoticed by the common Internet user. Email has been under a lot of pressure with more spam and more systems. Mail providers have taken notice and are helping their customers screen out the bad looking mail.
DynECT Email Delivery gives you visibility into what’s going on with your email and how you can grow into becoming a more reputable sender. Like with the expectations on website uptime, ISPs and anti-spam groups (and frankly, us) expect you to send email to those that are asking to receive it. On the flip side, those customers are expecting to receive those emails and often will get frustrated when they don’t.
Going to Two: When One Data Center Isn’t Enough
In late-September, a hurricane came up the East Coast of the United States with the path of the storm cutting over two of our 17 global data centers. While we practice redundancy and would manage just fine if something did happen, it’s a luxury that not enough of our customers take advantage of.
Moving to multiple data centers or multiple availability zones takes a lot of work for seemingly little gain. A lot of your application has to be deconstructed so that it can be split up. Our Managed DNS product allows you to manage and monitor your traffic easily. We watch your externally facing services and if we notice that they go down, we’ll swap your DNS records and point people to your second site.
Even if you don’t have a fully second instance of your site or application, our traffic management services will let you failover to a different URL — whether that’s a Twitter feed with status updates or a static page with a “Oops, we’re down” message.
Like DNS and Email above, the expectation is that data is always available. There are simply too many cost-effective scenarios and options for it not to be.