DevOps and Docker Talk: Cloud Native Interviews and Tooling

The episode contains solid technical depth about backup architecture fundamentals - deduplication, encryption, compression, indexing, and the distinction between incremental and index-reference based snapshots. However, much of the conversation circles around the same core concepts repeatedly rather than introducing new insights per segment. The discussion of entropy-based ransomware detection and the 24TB-to-159GB compression ratio are genuinely valuable datapoints, but filler about general backup challenges and the host's personal experiences dilutes density.

“When you tackle that issue of how do you find the proper thing to restore in a fast way, you end up realizing that you have to develop some kind of database.”

“We have almost 15,000 cycle of backups that we did, snapshot that we did, on this machine. the logical size is 24 terabytes. So, we I have here a huge amount of data, but the space that we are using is only 159 gigabytes.”

The core insight about decoupling storage from backup technology through end-to-end encryption that still enables deduplication is genuinely novel and contrarian to traditional backup vendor approaches. The reference-based snapshot model avoiding chain-of-dependency corruption is solid. However, much of the framing (3-2-1 backups, ransomware posture, integration plugin ecosystems) recycles established backup thinking. The demo website's immutable read-only snapshot interface is interesting but not deeply explored.

“What we are enabling with this new protocol is being able to, you know, asking to every team, make your backup, push this backup to a third party, internal third party or external third party, And this third party will manage resilience without any knowledge about your data.”

“Because we are doing the encryption, the compression, and the deduplication at source, it means that, all around the path where the backups are going, they are already super optimized in terms of storage and in terms of space.”

Gilles brings a decade of R&D-backed expertise in building backup systems from first principles, demonstrating genuine technical depth in cryptography, storage optimization, and system design. Julian provides CEO-level perspective on product strategy and community development. Both are practitioners who have clearly shipped production-grade systems, not merely theorists. However, neither guest is a marquee-name operator scaling a breakout company, and the conversation occasionally lacks the pressure-tested urgency of founders in hypergrowth.

“Gilles did a lot of R&D, in the past 10 years on it”

“We are a team of people who have worked in security a lot. We have been facing specs a lot about crypto in banking and stuff.”

The episode includes concrete metrics (24TB logical → 159GB stored across 15,000 snapshots; 10,000 cycles without size increase) and named integrations (Google Drive, iCloud, Notion, OneDrive, S3, GitHub). However, most implementation details remain abstract: the exact API complexity, integration development timelines ("half an hour" for Google Cloud is mentioned once), and deployment patterns are discussed conceptually. Few specific customer examples, deployment architectures, or comparative benchmarks are provided. The host's personal anecdotes (government enterprise, 7,000 users, 15 years ago) are vivid but not current evidence.

“the Google Cloud integration was done in half an hour, unplanned.”

“the logical size is 24 terabytes. So, we I have here a huge amount of data, but the space that we are using is only 159 gigabytes.”

The host asks solid follow-up questions about Docker deployment challenges, Kubernetes integration strategy, and the specific mechanics of recovery and deduplication. He pushes back on the registry-as-storage idea with practical objections and shares relevant operational context. However, the conversation often drifts into the host's personal anecdotes (hurricane recovery, ransomware monitoring) that occupy airtime without deepening the Plakar discussion. Few moments of genuine disagreement or pressure-testing emerge; most exchanges are affirming. The host rarely challenges claims or asks for comparative positioning against established vendors.

“So sign me up. I'm going to make you laugh. two days ago, I was having my sleepless night of wondering what I was going to do.”

“I'm here for the Docker backups. I'm here for, I'm here for, image registries to be a, from and to.”