Presented By O’Reilly and Cloudera
Make Data Work
21–22 May 2018: Training
22–24 May 2018: Tutorials & Conference
London, UK

Practical advice for driving down the cost of cloud big data platforms

Christopher Royles (Cloudera)
17:2518:05 Wednesday, 23 May 2018

Who is this presentation for?

System and Technical Architects, Program delivery, CTO, CTO

Prerequisite knowledge

General understandin of Cloud and Big Data will be useful Planning a large scale cloud deployment

What you'll learn

The audience will learn from practical and real-world experience in the initial stages and the longer strategic considerations of big data on cloud deployments. We will also introduce some of the sizing and budget considerations that ensure a project is successful through life.


Many organisations understand the value of Cloud and also the value of Big Data. There is a trend in addressing both the Cloud and Big Data use cases as a combined project. These projects may start with a modest investment but can scale very quickly with overwhelming demand from the business and seemingly exponential increase in data volumes. The early stage of all projects requires an appropriate kickstart approach and seed investment. Understanding how to execute these initial pilots, while having a view on a 5 year TCO can be challenging for organisations with limited experience of either cloud or big data technologies.

Successful projects maintain flexibility in deployment patterns, this enables tuning for workloads, tuning for data and continued optimisation of resource use and costs. Driving projects from user-need, through a use-case catalogue can assist in managing the data ingest, management, governance and exploitation of data. The use-case catalogue also assists in managing changing priorities and feeds into budgets and financial plans. Based on a range of successful production deployments on AWS and Azure, this session will provide the foundations of the MVP and kickstarting your project, it will then discuss how growth can be planned, managed and budgeted effectively.

Technical practitioners will learn how to effectively size an initial MVP, as well as pragmatic methods of scaling to a 5 year long term foundation and growth plan. Business sponsors and program delivery will benefit from a pragmatic retrospective on lessons learned and techniques for managing the initial funding and longer term strategic budget and investment.
This session will include the high level themes:

  • The technical components used in efficient cloud deployments
  • Wider discussion of options for use case and workload based patterns
  • How to scale the operational aspects
  • How to manage costs and reduce spend throughout the project

Sizing and effective planning are one aspect of a successfully project, it is also dependant on the combination of Cloud and software components to deliver efficiencies in data storage and analysis.
More detailed topics will include:

  • Use-case and Workload based patterns
  • Shared Storage
  • Scalable Analytics
  • Elastic Scaling
  • Hybrid and Cross Cloud Considerations
  • Patterns for Cloud Deployment
  • Tactical v Strategic Financial and Operational Planning
  • Cost Optimisation
Photo of Christopher Royles

Christopher Royles


Dr Christopher Royles is a Systems Engineer at Cloudera.

He holds a PhD in Artificial Intelligence from Liverpool University which he subsequently applied to voice recognition and voice dialogue systems. Chris has advised on Government Open Data initiatives as part of the Open Data User Group (ODUG) and sat on the quick wins stream of the UK Government Cloud Program (GCloud).

Chris has built out large scale data lakes on Amazon and Azure as well as assisted customers in their initial MVP through to full scale production.

Leave a Comment or Question

Help us make this conference the best it can be for you. Have questions you'd like this speaker to address? Suggestions for issues that deserve extra attention? Feedback that you'd like to share with the speaker and other attendees?

Join the conversation here (requires login)