December 8, 2019

369 words 2 mins read

Improving DevOps and QA efficiency using machine learning and NLP methods

Improving DevOps and QA efficiency using machine learning and NLP methods

DevOps and QA engineers spend a significant amount of time investigating reoccurring issues. These issues are often represented by large configuration and log files, so the process of investigating whether two issues are duplicates can be a very tedious task. Ran Taig and Omer Sagi outline a solution that leverages NLP and machine learning algorithms to automatically identify duplicate issues.
Talk Title Improving DevOps and QA efficiency using machine learning and NLP methods
Speakers Ran Taig (Dell), Omer Sagi (Dell)
Conference Strata Data Conference
Conf Tag Making Data Work
Location London, United Kingdom
Date May 22-24, 2018
URL Talk Page
Slides Talk Slides
Video

DevOps and QA issues usually consist of large files with numerous log events and configuration data. Resolving such issues is sometimes like finding a needle in a haystack. It requires the execution of highly technical procedures by experienced DevOps and QA engineers. When the issue is a duplication of an existing issue (for example, an additional failure of a certain test), this tedious investigation process may be avoided. However, identifying such duplications is a complicated task that depends on the investigator’s familiarity with past issues, knowledge sharing within the team, and thorough investigation of candidate issues. Software tracking systems (like JIRA or Bugzilla) typically enable textual querying for locating items of interest (log content, system documentations, configuration properties, labels, etc.). These search tools assume that data is of high quality and that user descriptions are semantically accurate. In reality, these conditions are not met and the investigation becomes a frustrating and time consuming task. Ran Taig and Omer Sagi outline a solution that leverages NLP and machine learning algorithms to automatically identify duplicate issues. The solution creates a “fingerprint” vector representation for each issue and stores each ‘fingerprint’ in a designated knowledge base. When a new issue arrives, its configuration and log files are processed through a pipeline that converts them into a new fingerprint. Recommendations for potential duplicates can be drawn using the designated knowledge base and a machine learning algorithm that aims to find similar fingerprints. Ran and Omer describe the successful implementation of this solution in Dell’s production systems, which has led to a significant reduction in the resolution time for issues.

algorithm tracking machine learning pipeline devops ran nlp
comments powered by Disqus
Apache Spark programming

Apache Spark programming

November 29, 2019

Brooke Wenig walks you through the core APIs for using Spark, fundamental mechanisms and basic internals of the framework, SQL and other high-level data access tools, and Sparks streaming capabilities and machine learning APIs.
Not your parents' machine learning: How to ship an XGBoost churn prediction app in under four weeks

Not your parents' machine learning: How to ship an XGBoost churn prediction app in under four weeks

November 22, 2019

Machine learning is a pivotal technology. However, bringing an ML application to life often requires overcoming bottlenecks not just in the model code but in operationalizing the end-to-end system itself. Goodman Gu shares a case study from a leading SaaS company that quickly and easily built, trained, optimized, and deployed an XGBoost churn prediction ML app at scale with Amazon SageMaker.
Human-in-the-loop data science with Jupyter widgets

Human-in-the-loop data science with Jupyter widgets

December 8, 2019

Jupyter widgets let you create lightweight, interactive graphical interfaces directly in Jupyter notebooks. Pascal Bugnion demonstrates how to use Jupyter widgets to implement human-in-the-loop machine learning with highly interactive user interfaces.
Humans and the machine: Machine learning in context (sponsored by IBM)

Humans and the machine: Machine learning in context (sponsored by IBM)

December 8, 2019

On the way to active analytics for business, we have to answer two big questions: What must happen to data before running machine learning algorithms, and how should machine learning output be used to generate actual business value? Jean-Franois Puget demonstrates the vital role of human context in answering those questions.
Lightning Talk: Scaling Distributed Deep Learning with Service Discovery: How CoreDNS Helps Distributed TensorFlow Tasks

Lightning Talk: Scaling Distributed Deep Learning with Service Discovery: How CoreDNS Helps Distributed TensorFlow Tasks

December 8, 2019

Training models with modern deep learning architecture is often computationally intensive and requires an efficient distributed system at scale. Such systems in distributed machine learning community …
Machine-learned model quality monitoring in fast data and streaming applications

Machine-learned model quality monitoring in fast data and streaming applications

December 7, 2019

Most machine learning algorithms are designed to work on stationary data, but real-life streaming data is rarely stationary. Models lose prediction accuracy over time if they are not retrained. Without model quality monitoring, retraining decisions are suboptimal and costly. Emre Velipasaoglu reviews monitoring methods, focusing on their applicability in fast data and streaming applications.