VoltDB and In-Memory Databases with John Hugg
Podcast: Play in new window | Download
Subscribe: Apple Podcasts | RSS
“There’s a lot of value in moving logic to the data rather than moving data to the logic. And the issue here is the data is a lot bigger than the logic.”
NewSQL is a class of modern relational databases that seek to provide the same scalable performance of NoSQL systems for OLTP, while still maintaining the ACID guarantees of a traditional database system. VoltDB is a NewSQL database, designed in part by Turing Award Winner Michael Stonebraker.
Today’s guest is John Hugg, a founding engineer at VoltDB. Today’s show covers the internals of VoltDB, including concepts like tunable consistency, database replication, and the advantages of a deterministic logical log.
- What is unique about VoltDB?
- Could you give some examples of where VoltDB would be a good fit?
- What are the design challenges of embedding application logic into a database?
- What are the ways developers can typically set up replication on their database?
- In the context of databases, what does determinism mean?
- In practice, what are the difficulties of implementing determinism on a distributed database?
- As you’re developing a database, what types of tests do you run to model customer use cases?
- How should systems architects decide which database to choose for their data management requirements?
- Michael Stonebraker
- All In With Determinism for Performance and Testing in Distributed Systems
- ECC memory
- Deterministic and Nondeterministic Functions
- “Taming the 9s” by Ariel Weisberg
- John on Twitter