Skip to content

Instantly share code, notes, and snippets.

@Semant1ka
Last active April 22, 2019 00:34
Show Gist options
  • Save Semant1ka/7f6a18283fa4227892d889048a07a6b3 to your computer and use it in GitHub Desktop.
Save Semant1ka/7f6a18283fa4227892d889048a07a6b3 to your computer and use it in GitHub Desktop.
simple_data_solution_comparison
Stack Performance Cost Usage Maintenance Scalability
Elasticsearch
Kibana dashboard
Depends on amount or RAM for cluster, if full index is RAM will have close to zero latency, also supports storing similar information in one cluster to enhance speed Elastic Only - $265.63/month
Elastic + Kibana by ElasticCloud on Google Cloud -$1464.16/month
Free trial available
Needs a special query language. Has beta SQL intefrace. Will be realized this year, beta is available now for usage. Easy Scalable - each node store some indexes, scaled by adding nodes.
MongoDb
BI connector
Visuablization
Free for Atlas users. Build to support SQL. Easy Not That Scalable, still needs SQL like schema, can’t do joins between collections."Scalability - if your queries will bring back millions of rows the BI Connector (an ODBC driver) will fall over. You'll need to do something on the software or hardware side to chunk the queries"
MongoDb
ETL
AWS redshift
Visualization
Pricing
No charge for data transfer between S3 and Redshift
Build on top of PostgreSQL, supports PostgreSQL syntax Somewhere in the middle. Scalable - scales into clusters, can set up clusters for different team, though there is a downtime when scaling occurred.
Hadoop
Spark
SQL database
Impala
Visualization
This is the most popular big data setup, query to hadoop itself can be slowed done when growing in size,so the best approach is to process data and store it in relational database. SQL query can be performed through Impala. Hard Scalable - can have as many nodes in HDFS as you want.
AWS Athena
AWS S3
$5.00 per TB of data scanned. SQL This solution looks like cutting edge no-ETL. AWS S3 is scalable noSQL storage
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment