Doing Data as a Frontend Developer
The contents here is created from the official CascadiaJS Page
Overview
- name: Tejas Kumar
- Speaker Page
Abstract
Web devs are often confused about data. What database do we choose? What are the tradeoffs between NoSQL and MySQL? This talk answers these questions and gives web developers DBA-level skills to help them build with data reliably.
Notes
From Data as a front end engineer
Go to text →
Data as a front end engineer
- Works at zeta
- Goal is to eliminate imposter syndrome
SQL or NoSQL?
… relational vs non relational
SQL
Pros
- Declarative
- Normalization - reduce redundancy
- Great for connecting data together
limitations
- Hard to scale
NoSQL
- Built for scale
- Heavily partitionable
- BASE - response to ACID
- Basically available soft state eventually consistent
NewSQL
- AWS aurora
- Convenience of sql with scale
You can use both… relational for current data Non relational for old archival data.
From Hunter
Go to text →
- adven-Tejas
- works at xeta
- SQL
- Has ACID interactions with the database
- Atomicity
- Consistency (TS at database level)
- Isolation
- Durability
- Has ACID interactions with the database
- Data Infrastructure > Database
- It's more important how you have things setup. You could move between a relational and a non-relational database.
- Search engine is part of data infrastructure.
- Tribalism is bad.
From Kevinlin
Go to text →
Notes by kevin
-
data in the browser, lots of variations
-
backend
- aws
- sql vs nosql
- relational vs non-relationsnal
-
relational & acid
- atomicity
- consistensity
- integrity
- data partition
-
limitation is scalability
-
nosql
- great for scale
- unstructured
- partionable
-
base
- basically available
- soft state
- eventual consistency
-
new sql
- sql & nosql
- eg. aurora, fauna
-
what to use
- it depends
- bank deposit: sql
- upvotes: nosql
-
why not both?
- data infrastructure > database
-
takeaways
- tribalism is bad
- integrity good
- data > databases
Children
Tags
Backlinks