DataStax Blog

#NoOutages- The world of the immortal online application

By Seema Haji -  May 15, 2014 | 0 Comments

Social media, mobile apps, email messages. These are all making it possible for customers to stay in contact with the companies they like most. Smart customers look for products, services and answers in many places and wherever it is most convenient. If your business cannot provide the service or information a customer wants, someone else will – friend or foe. In this always-connected world, are outages – planned or unplanned — acceptable? What if you never suffered an outage and customers could reach you at any time; how much more competitive would you be? How would that impact your brand and market share?

The Unplanned Outage – When a Natural Disaster Takes Out Your Business

Hurricane Sandy is the catastrophic test of how well businesses can keep going when a natural disaster strikes. The widespread outages caused by Sandy caught major organizations off-guard even with advance warning. ConEdison, an electric utility company serving New York City and Westchester County, left more than a million customers without power, after underground equipment was flooded and power lines came down. In Sandy’s immediate aftermath, more than nine days after the storm 79,000 ConEd customers were still without power. Sadly, ConEdison wasn’t alone: The country saw this story repeated over with many businesses caught in the brunt of Sandy as they grappled with overwhelmed disaster recovery plans.

Sandy taught us a valuable lesson; unplanned disasters can truly bring entire IT infrastructures to its knees.



Source: D&B.com

Enter Real-time Global Multi-Data Center Replication

Today’s modern businesses serving the smart customer require online applications to be available come hurricane or storm, literally. Some might argue that this magnitude of scale and disaster cannot be planned for. I disagree. In the past, companies relied on physically close data centers to replicate large amounts of data in real-time. Today, a workable disaster recovery plan needs a solution that spans multiple data centers and/or the cloud to ensure the fastest possible response times for customers and employees who are geographically separated and with no single point of failure. Technologies like Cassandra are leading this wave of change.

Cassandra is tailor made for this real-time and global multi-data center world. Cassandra’s peer-to-peer design that has no single points of failure makes it ideal for continuous availability and disaster avoidance. Because Cassandra is masterless, all nodes are equal, regardless of where those nodes are physically located. Cassandra can easily support a single database spanning multiple data centers where each center holds its own copy of the data and can have as many nodes as needed for supporting that site. If you think this is overkill, run a Google search on the term: “Hurricane Sandy Business Continuity.”

The Planned Outage – Maintenance, Upgrades and More

What’s ironic is that unplanned downtime makes up only a small proportion of the total downtime experienced by a typical company. Organizations routinely bring their servers offline on purpose for maintenance, upgrades and more. Planned outages are business as usual for IT. But today is a different age — an age of always on, instant connectivity, and global customer reach. Today, slow is down. And if your application is down, then your business is definitely out. Take Evernote’s planned outage for example. Three hours of downtime during peak business hours equals thousands of disgruntled customers.

Living in the World of the Immortal Online Application

We’re living in the world of the immortal online application where slow, unresponsive sites and downtime are unacceptable – planned or unplanned. While organizations may plan for disaster recovery, disaster avoidance built on an infrastructure that supports scale and prevents outages is key to the future of a business. Companies like DataStax help organizations innovate and scale online applications on an infrastructure that never fails. With built-in multi-data center and cloud replication and no single point of failure, Cassandra guarantees that your online data has zero chance of downtime whether outages are planned, unplanned, or both.

Use Case in Point – Outbrain on Cassandra during Hurricane Sandy.

Outbrain serves online content for some of the world’s largest media brands including Reuters, Wall Street Journal and USA Today and pushes out 90 billion recommendations on more than 10 billion page views per month. Outbrain relies on Cassandra as its massively scalable data store capable of handling high data velocities with 58,000 links to content per second. During Hurricane Sandy, Outbrain completely lost an entire data center but their data in Cassandra never went offline.

New to multi-data center? Download our white paper: “An Introduction to Multi-data center replication with Apache Cassandra, Hadoop and Solr.”



Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>