and on Substack

 and on Techgeezer Substack

Simulate Availability Zone failures on AWS

Deploying business-critical systems to Amazon Web Services (AWS), be prepared for anything, including an outage. The design needs to be at least highly available and, at the most, fault-tolerant. If the application or system goes down, there is the possibility of minimal interruption. In the context of AWS, regions and availability zones are used to build both highly available and fault-tolerant systems. But how do you plan if the unthinkable happens? (also on https://techgeezer.io)  

https://dzone.com/articles/chaos-engineering-simulate-az-failures-on-aws 

 #chaosengineering #chaosgroup #infrastructure #datacenter #amazon #aws #techgeezerio

Chaos Engineering

"Chaos Engineering is the discipline of experimenting on a distributed system in order to build confidence in the system’s capability to withstand turbulent conditions in production." -Chaos Monkey(gremlin.com)

https://www.gremlin.com/chaos-monkey/ 

 #chaosengineering #chaosgroup #embracethejourney #engineers #informationtechnology

Tesla Model,X hacked!

 https://www.embedded.com/tesla-model-x-hacked-with-195-raspberry-pi-based-board/?fbclid=IwAR0qIEPHTpGEm6_TzMPvrkY0BF63uQWmr71D90mYCxh-jbXIz4QRUg3C1EY

File database for Rapid Prototype

Using a JSON file DB is an exceptionally useful way to prototype but also is very expeditious. Have some way to rapidly deploy the data that arrives in its native format and even converting it to JSON is quite simple. IF you don't have a MongoDB instance handy, this is a spectacular solution. 

https://www.sitepoint.com/database-json-file 

#json #database #filedb #golang #mongodb #api #python3 #python #programming #datascience #techgeezerio

Adapting on the Fly to Test Time Distribution Shift

This is a REALLY good article on M/L for handwriting transcription. We did a similar technique for this but more along the track of image processing with Markov Chain/LSTM.

https://bairblog.github.io/2020/11/05/arm/ 

#ai #mlearning #mlengineer #techgeezerio

Clustering Algorithms

Clustering is a technique that most Data Scientists will use, especially in their journey toward Machine Learning (ML). When given a set of data points, clustering involves grouping them to use a selected algorithm to classify the data into a specific group. Theoretically, data points that share a "grouping" should have similar properties and features, while data points in a "different" group should have highly "dissimilar" properties or characteristics. If we consider cluster, it is an unsupervised learning method and a typical statistical data analysis technique used in numerous domains. Data Scientists can use clustering analysis to gain valuable insights from the data by understanding the groups that data points collect into when applying a clustering algorithm. 

Read the rest here.....

https://www.linkedin.com/pulse/clustering-algorithms-marc-buraczynski/

Test