---
title: "#187: SQL Keys in Depth"
publication: "DB Weekly"
issue: 187
published_at: "2018-01-19"
canonical_url: "https://dbweekly.com/issues/187"
---

Database Weekly 187   

This week's database news -[Read this on the Web](http://dbweekly.com/issues/187) 

[Database Weekly](http://dbweekly.com) Issue 187 — January 19, 2018     

[Datasette Publish: A Webapp for Publishing CSV Files as an Online Database](https://simonwillison.net/2018/Jan/17/datasette-publish/) — A tool that turns one or more CSV files into an online database with a JSON API, powered by SQLite and the [Zeit Now](https://zeit.co/now) deployment platform. Simon Willisontools [Bistro: A Lightweight Column-Oriented Data Processing Engine](https://github.com/asavinov/bistro) — Written in Java, Bistro is an interesting new light-weight column-oriented data processing engine based on a novel column-oriented logical data model. Alexandr Savinovcode [Use SQL in MongoDB? But of Course You Can. We'll Show You How](https://studio3t.com/download/?utm_source=Newsletter&utm_medium=DBWeekly&utm_campaign=Jan2018) — And there's so much more to discover. Like SQL Import & Export. Or our innovative Query Code that lets you instantly generate code from our Query Builder in Node.js, Python, or C#. Studio 3Tsponsored [![\](https://copm.s3.amazonaws.com/1817c866.png)](https://studio3t.com/download/?utm_source=Newsletter&utm_medium=DBWeekly&utm_campaign=Jan2018) [SQL Keys in Depth](https://begriffs.com/posts/2018-01-01-sql-keys-in-depth.html) — An attempt at boiling down the various arguments, pros and cons of different types of keys, whether they’re auto-incrementing integers, UUIDs, or other things. Joe Nelsontutorial [Fueled by Kafka, Stream Processing Poised for Growth](https://www.datanami.com/2018/01/18/fueled-kafka-stream-processing-poised-growth/) — Once a niche technique used only by large organizations, stream processing has emerged as a popular way to deal with massive amounts of freshly generated data. Datanaminews [How Alibaba Created a Stream System with Millions of Feeds](https://medium.com/@alitech_2017/how-to-create-a-stream-system-with-tens-of-millions-of-feeds-6df06e302c78) — A first hand look at how Chinese ecommerce company Alibaba scales using its own custom extensions to PostgreSQL. Alibaba Techstory [Using Redis as a True Distributed, Fault-Tolerant Database](https://blog.yugabyte.com/using-redis-as-a-true-distributed-fault-tolerant-database-8504a924a3f7) — The ‘trick’ here is that you’re not using Redis _per se_, but [Yugabyte](https://thenewstack.io/yugabyte-new-database-solve-sql-vs-nosql-dilemma/), a database that _supports_ the Redis protocol and can be used in-place of it. Karthik Ranganathantutorial [Go Deep into Today’s Modern Data Tier to See the Root of the Problem Fast](https://selectstar.io/root/?utm_source=CooperPress&utm_medium=Brief+Link&utm_campaign=Database+Weekly+Brief+Link+Jan19) SelectStarsponsored [How We Sped Up Eventually-Consistent Database Global Syncs with Caching](https://www.pubnub.com/blog/how-we-sped-up-eventually-consistent-databases-global-syncs-through-caching/) — They reduced an already solid time of 1 second for global consistency to 0.25 seconds by adding a volatile cache. Chaitanya Patchavastory [Don't Throw Out Your Algorithms Book Just Yet: Classical Data Structures That Can Outperform Learned Indexes](http://dawn.cs.stanford.edu/2018/01/11/index-baselines/) — An interesting response to [this paper](https://arxiv.org/abs/1712.01208) we linked to last year about machine learning based indexing systems. Bailis, Sheng Tai, Thaker, and Zahariastory [Graphile: Instant GraphQL APIs for Postgres Databases](https://www.graphile.org/) — Auto-discovers tables, columns, relations, procedures and more, then runs a high-performance secure GraphQL API server that adheres to best practices. Benjie Gillamtools [What's The Effect of 'Meltdown' on Postgres's Performance?](https://blog.2ndquadrant.com/postgresql-meltdown/) — How have the fixes for the recent Spectre and Meltdown CPU vulnerabilities affected Postgres performance? This first look suggests about a 7% hit. 2ndQuadrantnews [How to Migrate an Azure SQL Database to Amazon Aurora](https://aws.amazon.com/blogs/database/migrate-microsoft-azure-sql-database-to-amazon-aurora/) Steve Abrahamtutorial [Using SymmetricDS to Replicate Your Database Structure](https://medium.com/data-weekly/using-symmetricds-to-replicate-your-database-structure-c502240f9523) — An open source database replication tool that can be used to keep heterogeneous databases in sync. Chris Hensontutorial [Analyzing Cassandra Performance with Flame Graphs](http://thelastpickle.com/blog/2018/01/16/cassandra-flame-graphs.html) Jon Haddadtutorial [Linux Cloud Hosting Starting at 1GB of RAM for $5/mo](https://welcome.linode.com/db-weekly/?utm_source=cooperpress&utm_medium=email&utm_term=db&utm_campaign=db-weekly-brief) Linode Cloud Hostingsponsored [Scaling YugaByte DB to Millions of Reads and Writes](https://blog.yugabyte.com/scaling-yugabyte-db-to-millions-of-reads-and-writes-fb86cea5ff15) — An open source database that supports Cassandra and Redis APIs. Bharat Baddepudistory [PL/Swift: Write Custom PostgreSQL Functions in Swift](http://www.alwaysrightinstitute.com/plswift/) The Always Right Institutetools    

Database Weekly is curated by Peter Cooper [Read this issue on the Web](http://dbweekly.com/issues/187) [See our archives](http://dbweekly.com/issues) [Change your email address](*|UPDATE_PROFILE|*)  [Stop getting this newsletter](*|UNSUB|*)  

Published by [Cooperpress](https://cooperpress.com/)
Fairfield Enterprise Centre, Lincoln Way, Louth, LN11 0LS, UK