Recent high-profile outages shine spotlight on Availability requirements

John Metzger, VP of Global Product Marketing

Published date: May 10, 2017

It’s incredible to reflect on how much we rely on online services today. From paying bills to checking public transport timetables, we now place an ever-increasing amount of trust in data, and perhaps more importantly, access to that data.

In the last few weeks, we’ve seen some high-profile outages — from Amazon and Instagram, to WhatsApp.

Amazon suffered a significant outage that left its S3 service offline for hours. Amazon S3 underpins many AWS services and works as an object and media store for many other internet services that are relied on by millions every day. You can read a little more about the specific issue here, but an internal investigation determined human error was to blame for taking the service out of action. Regardless of the outage’s root cause, it resulted in a cascade effect that rendered large swathes of internet services unavailable for a period of hours, grinding businesses to a halt.

Last week, Instagram also suffered an outage, sparking a wave of Twitter posts from users lamenting the lack of access. While the exact cause of the downtime hasn’t been released, Instagram has one of the most scalable application architectures on the planet.

Instagram reached 700 million users at the end of April, meaning the company managed to double its user base in just two years. The platform serves up a daily total of 80 million photos and videos.

Web-based businesses have an implicit (or, in many cases, explicit) obligation to be available 24.7.365. Instagram powers countless businesses that rely on the digital platform to deliver ad views and content to customers, also setting the expectation of a service that will be Always-On. The cost to the many businesses that rely on Instagram serving content is not easily measured, but it is clear that Instagram suffered a reputational black eye during the time its services were unavailable.

WhatsApp is another tech giant recently kicked offline in a global outage, with some of its 1.2 billion users unable to use the messaging service for several hours.

At the end of the day, regardless of how scalable, resilient and distributed a system architecture is, anything that’s software driven and underpinned by application logic has the potential to break and suffer downtime. There is always an Availability Gap between services going down and being restored, which in turn has a knock-on effect for partners and customers alike. The 2017 Veeam Availability Report shows that 82% of enterprises across the globe cannot meet business requirements, which translates into costs of $22 million per year for the average enterprise.

This demands an Availability, protection and recoverability plan that’s separate from inbuilt logic. In today’s world, where consumers are enraged by downtime, reputations are made by ensuring that services remain available around the clock.

As businesses increasingly rely on the cloud to deliver services, any downtime has simply become unacceptable. At Veeam, we’ve addressed this concern by engineering a comprehensive platform that focuses on Availability for the Always-On Enterprise.

We’ll showcase many of our solutions at VeeamON 2017, THE Availability event of the year where you can learn how Veeam helps enterprises — those who are cloud native and those who have embraced digital — stay Always-On.

Interested in attending VeeamON 2017 in New Orleans, May 16 - 18? Go to our website and register. Hurry, seats are limited!

October 10th

3 Essential characteristics of managing planned downtime

October 5th

Veeam and NetApp: A partnership forged by change

September 27th

“Stacks” of Innovation from Veeam and Microsoft

September 25th

Ten years of the iPhone: Where are we today?