Exploring ChatGPT for Improved Observability

Overview

Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!

Grab it

Discover how ChatGPT can enhance observability in this 23-minute conference talk from Conf42 Observability 2023. Explore the importance of observability in modern ephemeral platforms and the rising costs of unexpected downtime. Learn about current observability challenges and innovative solutions, including the application of AI and large language models. Delve into the functionality of deep neural networks and their pre-training process. Examine ChatGPT's capabilities, limitations, and potential use cases in observability. Gain insights into prompt engineering as a new discipline and the significant investments being made in GPT technology by hyperscalers. Evaluate whether large models are a panacea for observability challenges and consider Dynatrace's approach to cloud management.

Syllabus

intro
preface
why do we care about observability?
modern platforms are ephemeral
cost of unexpected downtime has risen significantly
todays observability challenges
risk of it outages are set to grow even further in 2023
modern observability solutions
there is currently a lot of innovation occuring in ai, in particular llm's
imitate brain-like functionality using deep neural networks
dnn models are pre-trained from "the whole internet"
chat gpt
chat gpt is built on top of latest breakthroughs in language model design
chat gpt ≠ skynet
use cases of observability
keep in mid that chat gpt's responses are non-deterministic
chat gpt's informed decision making
prompt engineering is a new discipline
hyperscalers are investing heavily into gpt's
so are large models a panacea?
dynatrace - cloud done right