- Amazon’s cloud infrastructure has gotten to a point in both compute and managed services that most of the new rollouts will trend towards higher level services. Products like Redshift Spectrum (for running super power big data queries against a RedShift data warehouse and/or an S3 data lake) is an example.
- Amazon is continuing to court those that want to migrate infrastructure and SaaS ISVs, but they are also spending significant effort to make native cloud application development easier on their cloud platform. Take, for example, their new CodeStar product that sits on top of the infrastructure services (such as Code Pipeline) that they have already built. According to Amazon CTO Werner Vogels, their goal is to help developers, “Just write code and deploy it.”
- Most companies in the process of cloud migration still see the cloud as a “better data center.” For applications, they are taking very cautious steps into cloud adoption – keeping the most critical business applications well away from the cloud.
- The cloud is seen as a way to unshackle your developers and systems staff from procurement. Once the cloud has become a valid option, changes to infrastructure can happen in hours. This flexibility reduces the need for procurement to negotiate things like 5-year hardware and support contracts. The same is happening with licensed Saas software.
- The cloud has not fixed all of the challenges associated with “downtime.” Even if the downtime of a cloud service is <6 hours per year, if it happens during a key business time (say, during Cyber Monday), then the cloud will feel like a failure. It also does not help with poor software architecture and code; and thus, this remains largely a software architecture problem.
- Cloud databases, and their related technologies, are well on their way to greatly reducing the need for traditional databases and database licensing. You can expect traditional databases like Oracle and SQL Server to survive, but the landscape is looking very similar to the loss of landlines to cell phones and VOIP.
- Cloud computing, and new cloud services, continue to fix common application and infrastructure pain points. Things like “No servers to provision or manage;” “Scales with usage;” “Never pay for idle;” and “Availability and fault tolerance built in” should mostly disappear with a cloud native methodology.
- The largest barrier to cloud adoption at most companies is cultural. Education and adoption by technical and development staff is an important strategy for all the cloud vendors, and events like the AWS Summit SF 2017 are a good place to learn.
If you'd like to explore how nearly 20 years of custom software development experience applies to your application migration, don’t hesitate to connect.