AWS Summit: moving up the stack and technology convergence
Our AWS Practice Lead Guillaume Jaudouin attended the latest AWS Sydney Summit last week. Read on to hear Guillaume’s key takeaways from this event.
As is becoming usual, the AWS Sydney Summit was packed (6,000+ attendees), loud and intense. The keynote line up was impressive and the summit was the place to be for anyone wanting to catch up with the latest news in AWS.
Moving up the stack
AWS has accelerated its delivery of services in the SaaS area and it was mentioned throughout the summit that AWS’ focus is now further up the stack. This is not to say that PaaS and IaaS are to be taken for granted, nor that AWS will stop innovating in this area. After all, Security is still Day 0 of any project and new EC2 instance types are released all the time. However, AWS seems to be working more and more in the domains of Artificial Intelligence (Lex, Polly, Rekognition) and Machine Learning where they believe more business value can be added to their customers. A key factor for the success of these services will be how well they can integrate with the rest of the AWS offering in the long run, especially in the data processing and storage families.
AWS’ angle is somewhat different to other vendors as they approach software development from an infrastructure background while main competitors Azure and Google have been in the software game for longer and have more recently moved to IaaS and PaaS. One could argue that AWS has the strongest foundation to build on.
Data Technology Convergence
Until recently, there was a clear separation between traditional Data Warehouse on one side (Enterprise Data Warehouse, ETL…) and technologies related to what could be called Big Data (Hadoop, NoSQL…) on the other. On any project, one approach or the other was chosen.
We are now seeing many architectures that involve both traditional DW and big data paradigms. The Lambda Architecture is good example of this. Check the Altis AWS Practice Page for other examples of this technology convergence.
Another significant example is the upcoming fully managed ETL service, AWS Glue. A preview of this tool was given during the Summit and it appears that this tool will combine traditional techniques of ETL such as metadata discovery and graphical data transformation authoring with technologies usually found in the Big Data space: Python Scripting and Spark Containers for run-time transformation.
We are very keen to hear what your take is on this year’s AWS Summit!