Downtime is no Longer Acceptable

If you went to bestbuy.com and the site was unavailable, how long would it take for you to go to amazon.com or elsewhere to find what you wanted?  On average, it’s less than 30 seconds; it used to be much longer, but our society has grown impatient.  If you’re not available when customers are looking for you, they will move on.

When asked to evaluate different cloud providers, 80% of us consider cost, speed, management, capabilities, and scalability.  Notice anything missing?

Only 20% of us even trouble ourselves to address the issue of availability.  It’s axiomatic that none of those other things are relevant without availability.  Why is it so overlooked?

The reliability of online services has increased to a point that we now simply assume that whenever we go somewhere online, it will always be there when we need it. Even though you may not have personally witnessed an outage, consider these significant events:

You may not have noticed any of the above issues, but the reality is that outages happen daily. Naturally, larger well-recognized organizations receive the most attention when something hits the fan. But if the largest organizations suffer notable outages, there is no question it can happen to your organization too.

The Modern Age

While many companies still maintain a data center or server room on-premise, most have discovered the immense benefits of offsite servers.  The Cloud has made this so much easier.

If your company has an internal server environment, managing availability is a continuous task.  Your IT team probably works 24×7 tracking and optimizing uptime, trends in usage, life cycle management, and consumption.  The more complex the environment or application, the greater the skillset required on your IT team.

Take something as simple as file storage, for example. Consider the security of storing this information in multiple secure and remote sites, in an alternate geographic location. This could prove invaluable in the event of an office fire or natural disaster. In such an event, resumption of work may be as simple as logging onto your servers from any other internet connection.

Do you really need on-premise servers?

The Cloud simplifies IT infrastructure management for businesses of any size.  You can now obtain access to compute infrastructure, hosted software applications (SaaS) and eve raw infrastructure (IaaS) without having to invest significant capital. Often, it’s simply pay-as-you-go.

Cloud Computing also allows established businesses to obtain virtually maintenance-free off-site storage and processing power.  No need to build a remote data center, hire staff to maintain it, and pay for all the underlying infrastructure (like racks, cooling, UPS systems, generators and more) to keep everything running.

More Benefits

It is less costly

The cloud provider or commercial data center has their own staff, and they form part of the overall cost, inevitably less than what you would pay for an on-premise system.  Redundancy means that failing components in any system are instantly isolated and their work is picked up by an alternate component.  Proper architecture ensures that this occurs so quickly that it is imperceptible by humans.  Result: no downtime.

Distributing the load

If your headquarters is in Houston, but you do business online with Japan or even Europe, the latency for transferring large amounts of data back and forth can be frustrating, even if they are simply visiting your website.  This may be further complicated if servers reach peak capacity or internet connections are saturated – they can no longer respond in a timely manner.

Using distributed computing, or even a CDN, means that your customer in Japan interacts with a Japanese server with significantly reduced latency.  A locally up-to-date cache delivers the content directly, eliminating the lag and drastically improving accessibility.  Result: high speed continuous service.

Availability is King

Downtime is no longer tolerated today.  Historically, highly respected cloud providers have promised 99.99 percent uptime.  That translates to 52 minutes and 36 seconds per year of downtime. Surprising, isn’t it? At Total Uptime, we architected our network from the ground up to deliver 100% uptime. 100% uptime is beyond the capabilities of on-premise infrastructure, or even a single site data center provider, regardless of design.  This high level of availability demands an exceptionally stable environment, multiple points-of-presence around the world, and massive redundancy.

Uptime is what we do.  If you’ve not heard of Traffic Routing or Cloud Load Balancing before, it means that we direct visitor traffic to servers that can handle it.  If you have servers in cities A, B, and C, where cities A and B are overwhelmed with traffic, the excess can be redirected to the underused server in city C.  And with proper monitoring, traffic can be instantly re-routed in the event of an outage at any site.

Disaster Recovery

Of course, we spoke about natural disasters that could put you out of business, but there are other serious threats too.  Everybody has heard of Phishing, and knows that it represents a significant threat.  Stealing credentials and transferring money to offshore accounts cost companies billions last year.  But as the very wealthy like to point out, it’s only money.

A more serious threat is ransomware where the objective is still to get your money, but in this case they hold your information hostage.  Using a phishing technique they will work their way into your servers and encrypt everything, demanding a payment in untraceable crypto currencies such as Bitcoin.

Even more nefarious than that, criminals will walk around the parking lot of your company, dropping USB data sticks/thumb drives, or CDs/DVDs  with innocuous titles, in obvious spots so that employees will pick them up and then plug them into their workstation to view the contents.

In just a moment your entire server is encrypted, and your business comes to a grinding halt.  They give you a deadline for payment, advise when the decryption key will be destroyed and then even offer helpful advice and methods to aid you in transferring funds.

There are decryption techniques that can be effective, but it shouldn’t be necessary.  We’re certainly not going to pay these criminals, and we don’t have to, if we have redundant servers.  We simply erase the affected server and obtain a fresh copy from elsewhere on the network.  Access to an archival copy of your data makes restoration an inconvenience, not a business destroying threat.

If a customer-facing, web-accessible server becomes unavailable, the Total Uptime load balancer can simply reroute or redirect traffic from one data center to another for disaster recovery. Criminals collect millions of dollars every year from companies and individuals that don’t protect themselves.  It’s important to make sure that you are not numbered among their victims.

If downtime typically costs $5,600 to $9,000 per minute for companies that are reliant on digital data, can you really afford to be a victim?  Based on their profits for 2016, downtime for AWS works out to over $60,000 per minute.  Think in terms of essential data loss, productivity loss for your employees, incomplete transaction losses, and opportunities lost.

That is not the full extent however because there is damage to equipment, legal or regulatory repercussions, damage to your reputation among your customers or potential customers, and you stand a good chance of losing the confidence among your stakeholders.

The Takeaway

Downtime is simply no longer acceptable at any level.  You can’t tolerate it within your organization, and customers won’t tolerate it online.

It is said that 32% of downtime is caused by human error.  It might be lack of training, it might be a poorly written script or piece of code, it might be an employee succumbing to a phishing attempt.  The remaining 68% are generally driven by actual equipment failure or malice. How will you be affected?

Right now you have a chance to step out of that loop.  The more dependent we become on digital data, the higher the cost is going to be when something goes wrong.  Don’t you think you deserve 100% Total Uptime?