Any set of New Year’s resolutions should resolve to actually try and keep the resolutions. January 19th is known as “Quitter’s Day” because it takes at least 21 days to form a habit, and most people don’t make it past the 19th. We should all get a pass for 2020 as no one planned around a 100-year pandemic or the kind of massive change and disruption it brought to our modern world. With that said, 2021 should be a whole new beast. As an IT leader, it is a good time to take an in-depth look at how we handle our IT infrastructure and resolve to do better. 

  1. Get Cloud Costs under control. A significant amount of AWS’s profitability is “loose change” because companies waste a lot of their cloud spend. A recent report estimated that 30% of cloud spending is wasted and that this number is accelerating. There are many cost management solutions and best practices on the market. These best practices include granular usage telemetry, automated usage monitoring, elastic scaling based on thresholds, etc. Most of these require running your own VPC as multiple “as a service” vendors do not necessarily work well with cost control measures. Focus on the vendors that run in your own VPC and allow you control over size and cost.
  2. Get serious about security. Note: Cloud security is easily five pages on its own, but I’ve winnowed this down significantly for sake of brevity.  Anyone checking their credit alerts will notice that details are leaked daily. It seems like there is another story about a breach every week in the press. No one is being careless intentionally; leaks happen because security is a complex issue that requires a lot of attention to detail. Take a deeper look at your data at rest (databases, files, LDAP) and in motion (networks, WANs, ports). The challenge of security is to enforce reasonable rules without injecting too much complexity as complexity is where breaches live. In 2021, do not just run the standard audit but think through the company’s security architecture and security culture. Focus on simplifying the IT infrastructure as well as modernizing its security structure.
  3. Multicloud. The cloud oligopoly is like global warming. Everyone knows that at some point or other, it will directly threaten our livelihood. Most companies have already seen how cloud costs can get out of control. While standardizing on the services offered by AWS, GCP, or Azure may seem like a win-win at first, it sacrifices freedom in the long term, and the bill will still come due. Moreover, those services are generally offered at a premium over third-party equivalents. Aside from that, multicloud is a given for any global enterprise that needs to run services 24/7. No cloud vendor reaches into every region of the world, and multi-region outages have struck all existing vendors.
  4. Get Kubernetes complexity under control. Kubernetes has become ubiquitous among enterprise development teams, yet it is strangling many developers with 3000 line config files and week-long debugging sessions. Kubernetes needs to support multi-cloud, multi-region configurations with much lower complexity. Every organization should be not only putting their best and brightest into solving this for themselves but working to solve it as a global community.
  5. Pursue next-stage automation and analytics.  After years of investing in analytics, many companies do not have much to show for it. It just takes too long to get anything done, and business is too dynamic for the kind of baked answers that most MPP systems like Teradata, Greenplum, or Netezza are made to handle. Modern analytics are real-time. The time to deploy matters more than having perfect and efficient cubes. Cloud-based analytic systems that can operate on production data or stream it in real-time are becoming the new norm. If you are still batching, this is the year to look into speeding things up. Go real-time.
  6. Address Database Sprawl. Most data architectures are piecemeal. Often no one looks at the entire picture. Consider some consolidation based on what your real data needs are. Understandably, companies are reluctant to afford costly migrations but consider maintenance costs, capability needs, consistency, and lineage concerns. Look at whether each database both independently and as a piece of your larger architecture serves all of its purposes cost-effectively. Evaluate whether a migration might make sense considering all of the other costs and limitations.
  7. Prepare for edge computing. There are more devices and sensors, and mobile clients than ever before. There will be even more that we cannot foresee coming soon. If 2020 has taught us anything, it is that resilience should be built into every system. By moving more compute and storage closer to the client and replicated at the network’s edge, systems are more responsive and scalable, and more resilient. Edge computing takes the load off of centralized regional clouds, and in the event of an outage, more work can take place at the client and the edge.

2020 was a wild ride. Are you ready for 2021? Do you have a plan to make it the best year yet? What other New Years Resolutions are you adding for your organization and data infrastructure? 

 

Author

Posted by Ravi Mayuram, SVP, Products and Engineering and CTO

Ravi Mayuram is responsible for product development and delivery of Couchbase NoSQL offerings. He comes to Couchbase from Oracle where he was a senior director of engineering leading innovations in the areas of recommender systems and social graph, search and analytics, and lightweight client frameworks. He was also responsible for kickstarting the cloud collaboration platform. Previously in his career, Ravi has held senior technical and management positions at BEA, Siebel, Informix and HP in addition to couple of start ups including BroadBand office, a Kleiner Perkins funded venture. Ravi holds a MS in Mathematics from University of Delhi.

Leave a reply