EVENT - Tuesday
When6:00 PM - 9:00 PM
DSF Meetup with Qubole
Our next meetup is being supported by Cervello and Qubole and will be hosted at one of the most impressive rooftops in Shoreditch, The Ace Hotel. Aimed at data scientists and data experts, the interactive tech talks & whiteboard sessions will cover the make-up and experiences of how to build and utilise a Modern Data Architecture.
5:45-6:15 Mingling – beers & refreshments
6:15-6:30 Introductions & meet up group announcements
6:30-7:00 Talk #1 – Enabling Data Science through a Modern Data Architecture
7:00-7:15 Ask an Architect – grab a beer and pizza and join one of the many whiteboard sessions (data science / modern data architecture)
7:15-7:45 Talk #2 – 2018 Big Data Cloud Trends
7:45-8:15 Real-time Retail Data Science Pipeline Deep-Dive
8:15-9:00 Ask an Architect – grab a beer and pizza and join one of the many whiteboard sessions (data science / modern data architecture)
Here’s what you’ll take away from this meetup:
Best practices for building an end-to-end data operation in AWS from ingestion to reporting and data science
The constructs of a Modern Data Architecture (MDA) in the cloud without breaking the bank
Right tool right use case: Hadoop, Spark and Presto to advance your business insights – from real time analytics to machine learning and stream processing, while reducing costs on your data warehouse with proper ETL methodologies.
Optimization of data pipelines using EC2 Spot Instances and other automation technologies for big data
Session #1 – Enabling Data Science through a Modern Data Architecture
Glyn Heatley from Cervello will talk through the principles of a Modern Data Architecture (MDA) and about the different roles Qubole plays in MDA, from data ingestion pipelines, ETL to ML. The principles of a Modern Data Architecture take into consideration the demands of high growth, innovative, acquisitive organisation that is flexible, extendable and scalable. Qubole enables flexibility in a MDA that extends beyond the rigidity of the tenets of legacy data architecture solutions. Glyn will highlight how Qubole enables businesses to leverage both governed and semi-governed data to quickly derive insights and build complex models without building out huge on-premise enterprise data infrastructure.
Session #2 – 2018 Big Data Cloud Trends
Ajith Ramanath, Qubole’s Technical Director of Solutions Architecture EMEA, will dive into how different Data Engineering and Analytics teams use Hadoop, Presto and Spark and why you should look at these engines for different use cases (ML, ETL, concurrent interactive analysis, etc.). As well as industry trends that are coming out of these engines for improved reliability, performance, and cost for different workloads in the cloud. Ajith will highlight some examples in which data teams are leveraging Modern Data Architectures to consume, read and productionise information, as well as how Qubole is enabling them to achieve 70-90% in cost savings at petabyte-scale.