Reliability Engineering

Reliability management is a crucial aspect of manufacturing that aims to ensure that equipment and systems operate at the highest level of efficiency and effectiveness. By implementing reliability management practices, manufacturers can reduce downtime, improve production efficiency, and ultimately increase profitability. One key aspect of reliability management is preventative maintenance…

Preventing equipment failures is crucial for maintaining the reliability and efficiency of your assets. One powerful tool for achieving this is Root Cause Analysis (RCA). …

by Marina Huberman Our team at Emi uses Sentry as the main error-monitoring software, which helps us not only to diagnose and fix errors but also, allows triggering anomaly alerts by using warnings along with statistical alert rules. About Elastic Beanstalk Elastic Beanstalk is a great option for starting with microservices in AWS…

How we ensured reliability for customer data — The sibling of security is reliability. We are building an end-to-end encrypted photo storage service. In this post we describe how we ensure reliability by keeping 3 copies of customer’s encrypted data, spread across 3 different cloud providers. Threat model Data loss due to machine failure is already covered even with a…

What is Bloat The record that has been deleted but is still taking up some space is called a dead tuple. Once there is no dependency on those dead tuples with the already running transactions, the dead tuples are no longer needed. PostgreSQL runs VACUUM on such Tables. VACUUM reclaims the storage occupied…

Death by a thousand cuts by wimpy fishermen for malicious whales. Data availability sampling - I have been looking into the subject for quite a few months. I looked at so many perspectives and somehow never got the full picture as I had been missing the motivation for the solution…

This post covers some content from the “Infrastructure Quality” lectures of our Machine Learning in Production course. For other chapters see the table of content. As discussed throughout this book, machine learning contributes components to a software product. While we can test machine-learned models in isolation, can test individual steps…

If machine learning has Curse of Dimensionality [1], I personally feel that data pipeline is also cursed of data dependencies. Cursed is a strong word, but I feel the word represents data pipeline quite well, especially in terms of reliability. Why is that so? What to do about it to…