Statistical Aspects of Distributed Tracing | Splunk
Get the slides: https://www.datacouncil.ai/talks/statistical-aspects-of-distributed-tracing?hsLang=en
ABOUT THE TALK
The idea of distributed tracing is to stitch together the execution path traversed by a request: operations are timed, and the execution context is propagated as different services perform work to handle the request. As storing all trace data is prohibitively expensive, it is necessary to select certain traces to be retained and discard others. Anomalous traces are invaluable to debugging and optimization workflows, but traces do not announce up front whether they will take an abnormally long time to complete, or whether an operation 35 links away will result in an error. A tail-based approach, in which the decision whether to retain a trace is deferred until the trace is complete, at which point its characteristics determine the likelihood that it is retained, is therefore required.
This talk will describe the product and engineering requirements for a robust and scalable tail-based distributed tracing system, and the statistical techniques that arise in meeting these requirements. For example, we will discuss how to prefer abnormally long and/or erroneous traces while maintaining the ability to calculate accurate summary statistics.
We will introduce the necessary concepts from distributed tracing. Some comfort with statistical arguments would be helpful.
ABOUT THE SPEAKER
Joe Ross holds a PhD in mathematics from Columbia University and was a researcher and instructor in pure mathematics, most recently at the University of Southern California. He has given more than 20 talks about his research at conferences and universities throughout the world (Germany, Japan, Turkey, USA). He has also been the primary lecturer for many undergraduate and graduate math courses, and has given countless informal seminars. He has 9 publications in peer-reviewed mathematics journals. Joe has worked as a data scientist at machine learning/analytics startups for over five years; in his current role, he focuses on a variety of time series (anomaly detection, forecasting, correlation) and sampling problems that arise in monitoring.
ABOUT DATA COUNCIL:
Data Council (https://www.datacouncil.ai/) is a community and conference series that provides data professionals with the learning and networking opportunities they need to grow their careers. Make sure to subscribe to our channel for more videos, including DC_THURS, our series of live online interviews with leading data professionals from top open source projects and startups.
FOLLOW DATA COUNCIL:
Twitter: https://twitter.com/DataCouncilAI
LinkedIn: https://www.linkedin.com/company/datacouncil-ai
Facebook: https://www.facebook.com/datacouncilai
Eventbrite: https://www.eventbrite.com/o/data-council-30357384520
Видео Statistical Aspects of Distributed Tracing | Splunk канала Data Council
ABOUT THE TALK
The idea of distributed tracing is to stitch together the execution path traversed by a request: operations are timed, and the execution context is propagated as different services perform work to handle the request. As storing all trace data is prohibitively expensive, it is necessary to select certain traces to be retained and discard others. Anomalous traces are invaluable to debugging and optimization workflows, but traces do not announce up front whether they will take an abnormally long time to complete, or whether an operation 35 links away will result in an error. A tail-based approach, in which the decision whether to retain a trace is deferred until the trace is complete, at which point its characteristics determine the likelihood that it is retained, is therefore required.
This talk will describe the product and engineering requirements for a robust and scalable tail-based distributed tracing system, and the statistical techniques that arise in meeting these requirements. For example, we will discuss how to prefer abnormally long and/or erroneous traces while maintaining the ability to calculate accurate summary statistics.
We will introduce the necessary concepts from distributed tracing. Some comfort with statistical arguments would be helpful.
ABOUT THE SPEAKER
Joe Ross holds a PhD in mathematics from Columbia University and was a researcher and instructor in pure mathematics, most recently at the University of Southern California. He has given more than 20 talks about his research at conferences and universities throughout the world (Germany, Japan, Turkey, USA). He has also been the primary lecturer for many undergraduate and graduate math courses, and has given countless informal seminars. He has 9 publications in peer-reviewed mathematics journals. Joe has worked as a data scientist at machine learning/analytics startups for over five years; in his current role, he focuses on a variety of time series (anomaly detection, forecasting, correlation) and sampling problems that arise in monitoring.
ABOUT DATA COUNCIL:
Data Council (https://www.datacouncil.ai/) is a community and conference series that provides data professionals with the learning and networking opportunities they need to grow their careers. Make sure to subscribe to our channel for more videos, including DC_THURS, our series of live online interviews with leading data professionals from top open source projects and startups.
FOLLOW DATA COUNCIL:
Twitter: https://twitter.com/DataCouncilAI
LinkedIn: https://www.linkedin.com/company/datacouncil-ai
Facebook: https://www.facebook.com/datacouncilai
Eventbrite: https://www.eventbrite.com/o/data-council-30357384520
Видео Statistical Aspects of Distributed Tracing | Splunk канала Data Council
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
DC_THURS : dbt w/ Drew BaninDevOps for Machine Learning & Other Half Truths Processes & Tools for the ML Lifecycle | DataRobotData Discovery Getting More From Your MetadataTechnical Founders PanelFeed The Alligators With the Lights On: How Data Engineers Can See Who Really Uses Data | StemmaArchitecting a Low-Latency Schemaless SQL Engine | RocksetBuilding High Performance Recommender Systems with Feature Stores | TectonOffice Hours with Stitch Fix Data PlatformDC_THURS on TrinoEnterprise Data Science Comes of Age | AnacondaUsing Machine Learning and Observability Together to Reduce Incident Impact | DigitalOceanMaking Friends with Generative Models | TonicThe Right Way to Track Mobile DataDC_THURS on Feature EngineeringScaling Uber's Metric System from Elasticsearch to Pinot | UberRikai: A New Data Format for Analytics on Unstructured Data at ScaleDC_THURS on DataHub w/ Shirshanka Das (Acryl Data)The Road to Exceptional Data CorrectnessBuilding an ML Experimentation Platform for Easy Reproducibility | TreeverseHow Vercel Builds Dozens of Metrics from One Heterogenous TableDC_THURS w/ Patrick Thompson, CEO of Iteratively