Take one: A mysterious menace chases you through a house without exits. Cut! Take two: You are leading a corporate meeting in your pajamas. Cut! Take three: Your company's network crashes.
We all have nightmares. For database professionals, the monster lurking in the dark is downtime— whether caused by disaster, hacking, human error, or outage.
Considering Business Costs of Downtime
Information Technology Intelligence Consulting (ITIC), a digital industry analyst, reported in May 2017 that downtime losses average $100,000 per hour. However, ITIC stated that an hour of downtime for the largest companies (about 33% of those surveyed) costs about $1 million. It further noted that for some companies -- such as banks, stock exchanges and utilities — the losses may reach "millions of dollars per minute."
When an employee made a mistake at an Amazon AWS datacenter on February 28, 2017, the resulting outage lasted about four hours. This led to losses of $150 million for S&P 500 companies and $160 million for financial-service firms, according to Vanity Fair. The outage harmed more than half of top Internet retailers according to Business Insider.
Avoiding Nightmares with ObjectRocket
At ObjectRocket, we're on task and provide engineering support for our DBaaS (database as a service) customers around the clock. We never charge you a fee for support. It’s always included, no matter how much data you have on our platform.
We have monitoring set up for all of our customers for common issues such as instances running out of disk space or Elasticsearch clusters going into red status. That way, we’re on top of potential issues before our customers reach out and many times before our customers even notice.
Support from Real Engineers
We’re here 24x7x365. When you alert us to a problem, one of our engineers provides support. We don’t support our customers with call centers and a one-size-fits-all script. We provide real engineers who are familiar with your setup and can start working on the problem quickly. It's one of our many practices designed to minimize downtime and keep nightmares away.
Our 99.95 percent platform uptime guarantee can also help you sleep better. We offer geo-redundancy to deploy customer databases to datacenters in other regions before disasters strike. Our large system of datacenters allows us to shift workloads easily and smoothly so we’re always ready if disruptions occur.
ObjectRocket customers never have to fear a sudden growth in data. We're proactive. We always over-provision for storage, in case you need it. That way, you’re less likely to experience failover.
Some of our other procedures include Agile management, which includes immediate remediation of problems through an all-hands-on-deck response we call unified escalation.
Agile is an alternative to old-style waterfall management. Instead of dividing staff into groups with each team working separately on a portion of a problem or product and then cascading it on to the next group, Agile teams review and suggest improvements to each other's task (sprint) as the project progresses. On the rare occasion there is an issue, we're all running together and communicating rapidly to escalate a solution to your problem. Later, we use Jira to track action items that are identified during a post-incident retrospective to ensure that preventable problems never happen again.
ObjectRocket's management system and infrastructure greatly improve the odds of avoiding human error, surmounting disasters, and recovering rapidly from database failure. We can't guarantee sweet dreams, but we encourage them. Check out our offerings or email us to set up a chat about your data infrastructure and how we can help.