TL;DR
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
The Pragmatic Engineer published a podcast episode featuring Sam Newman discussing his new book, Building Resilient Distributed Systems, and practical choices for designing and operating distributed software. The conversation also examines microservices, observability, failure handling and how AI tools may affect developers’ understanding of systems.
The Pragmatic Engineer has published a podcast episode in which software architect and author Sam Newman discusses resilient distributed systems, microservices and the effects of AI on software development. The conversation introduces ideas from his new book, Building Resilient Distributed Systems, and focuses on practical engineering decisions that can affect service reliability and teams’ ability to change software independently.
Newman describes microservices as an architecture of “last resort,” despite having written Building Microservices, one of the best-known books on the subject. In the episode, he discusses why teams may choose the approach, what they can get wrong, and why independent deployment is a useful test of whether services are meaningfully separate. He also identifies business-domain boundaries as a softer way to define a microservice.
For distributed systems, Newman reduces a familiar set of risks to three principles: information takes time to travel, the resources a system depends on may be unavailable, and compute, memory, storage or network capacity can run out. He says resource exhaustion is behind most outages he has encountered. The episode also addresses observability and the business consequences of choosing whether a system should fail open or fail closed when errors occur.
The discussion covers idempotency, a property that prevents repeated requests from producing repeated effects. Newman uses payments as an example: a retry should not charge a customer twice. The source describes two approaches: clients can send idempotency keys, or servers can identify repeated requests using fingerprints. Keys require changes to both client and server, while fingerprints can be added on the server but may mistake legitimate requests for duplicates. The supplied source text cuts off before completing its account of the trade-offs.
Reliability Depends on Design Choices
The episode’s focus is relevant to engineering teams because distributed systems can fail for ordinary reasons: delays, unavailable dependencies and exhausted capacity. Newman’s emphasis on resource limits points to a practical reliability concern: systems need design and monitoring that account for finite resources, not only network or service outages.
His discussion of failure behavior also highlights that there is no single response suitable for every service. Whether a system should fail open or fail closed depends on the business consequences of allowing an operation to continue or stopping it. Likewise, microservices can support teams that need to release changes independently, but the episode frames that architecture as a consequential choice rather than a default target.
The AI discussion adds a team-level concern. The source says Newman considers cognitive debt and cognitive surrender, and argues that modular architecture can help teams try AI tools while retaining an understanding of the systems they build. These are topics raised in the conversation, not measured findings about AI’s effects.
distributed system monitoring tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Newman’s Microservices Background
Newman is closely associated with the development and popularization of microservices. According to the source report, he was present at an architecture symposium in England in the early 2010s where participants discussed naming this style of service-oriented architecture. James Lewis and Martin Fowler later published “Microservices: a definition of this new architectural term” in March 2014; Newman’s book Building Microservices followed in 2015.
The episode returns to the definition question. Newman’s clearer criterion is whether a service can be deployed independently after a change, without requiring other services to be deployed at the same time. His broader definition emphasizes boundaries around business functions, rather than splitting an application solely by technical layers. The podcast also draws on his work teaching automated testing and on his experience in consulting, startups and independent work, as presented in the source report.
““last resort””
— Sam Newman, as quoted in The Pragmatic Engineer’s episode report
As an affiliate, we earn on qualifying purchases.
Episode Details Still Missing
The source identifies the episode and its topics but does not provide a publication date, episode number or the full transcript. It says listeners can find the transcript near the top of the original page and timestamps at the bottom, but those details are not included in the supplied material.
The provided text also ends partway through its description of request fingerprints. It therefore does not establish the full set of limitations Newman discussed or offer enough detail to compare the two idempotency methods for particular systems. The reported observations about causes of outages are attributed to Newman’s experience; the source does not provide incident data or an independent analysis validating them.
As an affiliate, we earn on qualifying purchases.
Where to Hear the Discussion
The Pragmatic Engineer says the episode is available on YouTube, Apple and Spotify, with a transcript and timestamps on its webpage. Readers seeking Newman’s full explanations of failure handling, idempotency, observability and AI-assisted development can consult the complete conversation; the supplied report does not announce a further release date or follow-up event.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is the news?
The Pragmatic Engineer published a podcast episode featuring Sam Newman on resilient distributed systems, microservices and AI’s role in software development.
What does Newman mean by microservices?
His clearest test is whether a service can be deployed independently from the rest of the system. He also uses a broader definition based on services divided around business functions.
What are Newman’s three distributed-systems rules?
Information takes time to travel, dependent resources can become unavailable, and resources such as CPU, memory, storage and network capacity are finite.
Where can listeners find the episode?
The source says it is available on YouTube, Apple and Spotify. It also points to a transcript and episode timestamps on The Pragmatic Engineer’s page.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
