Distributed Data Show Episode 51
David and Jeff talk with Ted Wilmes from Expero about best practices regarding DSE Graph and the importance of proper data modeling.
0:16 - David goes through introductions and gets an answer about who/what is Expero
0:57 - Expero has the best swag and David realizes his laptop is already wearing some
2:13 - What does Expero help developers understand?
3:55 - Ted explains why core DSE data modeling is so important because everything you do in DSE Graph eventually maps down to your model
4:39 - Potential graph pitfalls from not understanding implications of the underlying data model
6:40 - We dig a little deeper into the how and why some models might cause larger and less performant partitions
10:04 - Temptation to take a logical (whiteboard) drawing and code it directly into the actual data model
11:15 - Ted talks to some other challenges with writing large datasets, data streams, and maintaining a production graph
12:59 - Finding a balance between read and write operations and potentially using a cache to reduce unnecessary reads
14:38 - Performance considerations when dealing with reads
17:04 - Best practices when using indexes in DSE Graph
19:45 - Understanding data model implications are required for performance at scale
21:23 - Parting words on the importance of checking out DataStax Academy and the value of watching some of the intro to Apache Cassandra videos
Developer Relations at DataStax