Shaping Big Data - Schema on Read or Schema on Write

Data Lakes often have the some of the same performance and security decisions as past year's data warehouses.  

Teams need to decide if the data in a lake is stored in producer formats or consumer formats or a combination of the two. 

Storage is essentially unlimited which means we may choose to store the data in multiple consumer oriented fashion.

Compute is essentially unlimited. We may decide to apply view style restrictions and access controls at read time.  		          

Video

Speaker's Notes

This discussion is really only about tabular style data stored in cloud blob/object stores.  See data lakes for squares. Record oriented data can be built up from files/objects that are organized in folders/partitions that are then aggregated into virtual tables.  

Data layout in the lake dramatically affects performance. Data objects are read in their entirety.  Joins are essentially external operators.  Views are not a native concept.  We may chose to organize the data based on the consumption patterns.

One option is to use the essentially unlimited compute to pull in data some format and filter the data as it streams to the client.  The client schema built as the data is read.

Schema on Read is often built by some type of proxy that sits between native cloud storage and the consumer.  See https://joe.blog.freemansoft.com/2020/05/fine-grained-controls-schema-on-read.html as an example of hos this could be implemented

Schema on Write means that the data is stored in the lake in format or formats that are targeted to the consumers or oriented towards simplifying attribute style filtering. Schema on write trades off space and potential data reorganizations against simplified security administration and the ability to use cloud native access controls.

The simplest version of Schema on Write is to create copies of the data for each consumer type.  A custom table is written to the cloud for each view pattern with a special emphasis on sensitive columns.  Consumers may end up with more columns than they need but never any they shouldn't see.  The example here has 3 tables, the original and two formatted to hide sensitive information from various consumers.

A slightly finer grained, more heavy handed approach is to break the data into groups. Consumers are given access to the appropriate column groups.  Essentially each of the column groups is its own table. The consumer reassembles the column groups to create a full table within the limits of their permissions. 

A totally cloud only approach is to break each data column into its own file/table.  We can take this approach because we are not constrained by normal DB stores or compute resources. This approach makes it possible to set permissions on individual columns using standard cloud access controls. Some data products built on cloud storage take this approach
Created 5/2020

Comments

Post a Comment

Popular posts from this blog

Installing the RNDIS driver on Windows 11 to use USB Raspberry Pi as network attached

Image
I do a lot of my development and configuration via ssh into my Raspberry Pi Zero over the RNDIS connection. Some models of the Raspberry PIs can be configured with  gadget drivers  that let the Raspberry pi emulate different devices when plugged into computers via USB. My favorite gadget  is the network profile that makes a Raspberry Pi look like an RNDIS-attached network device.  All types of network services travel over an RNDIS device without knowing it is a USB hardware connection. A Raspberry Pi shows up as a Remote NDIS (RNDIS) device when you plug the Pi into a PC or Mac via a USB cable. The gadget  in the Windows Device Manager picture shows this RNDIS Gadget connectivity between a Windows machine and a Raspberry Pi. The Problem Windows 11 and Windows 10 no longer auto-installs the RNDIS driver that makes magic happen. Windows recognizes that the Raspberry Pi is some type of generic USB COM device. Manually running W indows Update  or  Upd...
Read more

Understanding your WSL2 RAM and swap - Changing the default 50%-25%

Image
The Windows Subsystem for Linux operates as a virtual machine that can dynamically grow the amount of RAM to a maximum set at startup time.  Microsoft sets a default maximum RAM available to 50% of the physical memory and a swap-space that is 1/4 of the maximum WSL RAM.  You can scale those numbers up or down to allocate more or less RAM to the Linux instance. The first drawing shows the default WSL memory and swap space sizing. The images below show a developer machine that is running a dev environment in WSL2 and Docker Desktop.  Docker Desktop has two of its own WSL modules that need to be accounted for.  You can see that the memory would actually be oversubscribed, 3 x 50% if every VM used its maximum memory.  The actual amount of memory used is significantly smaller allowing every piece to fit. Click to Enlarge The second drawing shows the memory allocation on my 64GB laptop. WSL Linux defaul...
Read more

Almost PaaS Document Parsing with Tika and AWS Elastic Beanstalk

Image
The Apache Tika project provides a  library capable of parsing and extracting data and meta data from over 1000 file types.  Tika is available as a single jar file that can be included inside applications or as a deployable jar file that runs Tika as a standalone service. This blog describes deploying the Tika jar as an auto-scale service in Amazon AWS Elastic Beanstalk.  I selected Elastic Beanstalk because it supports jar based deployments without any real Infrastructure configuration. Elastic Beanstalk auto-scale should take care of scaling up and down for for the number of requests you get. Tika parses documents and extracts their text completely in memory. Tika was deployed for this blog using EC2 t2.micro instances available in the AWS free tier. t2.micro VMs are 1GB which means that you are restricted in document complexity and size. You would size your instances appropriately for your largest documents.   Preconditions An AWS account. AWS ac...
Read more