New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
Story of every modern project ever,
Start with MongoDB, Serverless Computing, Kafka, Kubernetes/Docker, Terraform, Consul, Ansible etc... spend 6 months on architecture design meetings, learning, installing and configuring and then realize you have no running app yet.
Then come to a conclusion that you could do the same in PHP+MySQL(or RoR or Django) in under a week and host it for $1/Month hosting.
If you haven't realized the above, you likely have a lot to learn.
I learned more here than I learned in my college.
Crazy talk, you must be on ACID.
Indeed. Postgres is really nice, and the community adds features as needed. Adding JSONB was especially nice. Instead of two different DBs I only have to deal with one.
Granted, I wasn't doing anything super complex, and I had a mixture of consistent data and really random data.
True, redis is mainly known for being used as a caching layer, but it's more then that though. It has PubSub features, and it stores several different data structures, which JSON is not one of then weirdly enough. Redis, the company, calls it a data structure database.
NoSQL generally means HTTP is used to interact with the database, or it uses some query language (QL) which isn't SQL. Postgres wasn't a SQL DB until 1995. It was a relational database, but it had it's own QL. Elastic has added SQL to Elasticsearch, so now Elasticsearch is an SQL database too.
It's mostly coincidental that most of the NoSQL databases also aren't relational. Columner databases like Cassandra have nothing in common with document databases like MongoDB, CouchDB, or Elasticsearch, or time series databases like InfluxDB or OpenTSDB.
NoSQL is also mostly a catchy marketing term, and it's generally pretty useless outside of marketing since it's such a catchall. There have been lots of different databases for ages. BerkleyDB, LMDB (which incidentally powers Redis), and others have been around for years before the NoSQL term was coined.
The search part varies. Elasticsearch is built for searching data inside documents. MongoDB is not.
I knew this fact but had forgotten it. You are quite right.
Got this shirt at Oracle OpenWorld a few years ago. It was for a company offering "transactional hadoop".