Scalability! But at what COST?
http://www.frankmcsherry.org/assets/COST.pdfA paper dissecting the notion of scalability and measurements of scalability in 'big data' systems. They define a metric, the Configuration that Outperforms a Single Thread, which is particularly aimed at giving meaningful comparisons, or: "without rewarding systems that bring substantial but parallelizable overheads".
There is a less academic overview on The Morning Paper: https://blog.acolyer.org/2015/06/05/scalability-but-at-what-cost/
Tags
Related By Tags
- ๐ Get Started ยท Snorkel
- ๐ Datasette โ Datasette documentation
- ๐ Computer Files Are Going Extinct - OneZero
- ๐ Command-line Tools can be 235x Faster than your Hadoop Cluster - Adam Drake
- ๐ Ibis: Python Data Analysis Productivity Framework โ Ibis 1.2.0+99.g7fc58a1 documentation
- ๐ Gazetteer of Australia 2010 - Datasets - data.gov.au
- ๐ [1903.03129] SLIDE : In Defense of Smart Algorithms over Hardware Acceleration for Large-Scale Deep Learning Systems
- ๐ What is dbt?
- ๐ All GDELT Event Files
- ๐ Becoming a Data Scientist (slides)
Details
- Revised
- Created
- Edited