Skip to content

Topic

Edge computing

Running application code at points of presence close to users rather than a single origin server, usually under tighter runtime and API constraints.

Current stories

buildOne report1 publisher

Cloudflare Gateway can now hand traffic to a customer's Worker on every Zero Trust plan

Cloudflare added a Forward action and custom validation functions to Gateway, putting customer Workers inside policy evaluation on all Zero Trust plans. Cloudflare says the hand-off is a local function call on the same machine. Operators still need to know what happens to a blocking decision when their code fails.

Reality

Evidence30
Adoption
Not scored
Hype gap+35
Incentives80
Confidence45
buildOne report1 publisher

Cloudflare designed D1's control plane to keep single-region components off the query path

Cloudflare says D1 has passed 5.5 million databases, with live databases up 15x in a year as of October 8, 2026. Its post lists the design tenets behind the control plane and says operators building on its platform can reuse many of them.

Reality

Evidence50
Adoption62
Hype gap+15
Incentives70
Confidence55
investConfirmed3 publishers

Omron agrees to pay $675 million for Onlogic, a US maker of industrial PCs

Omron will buy US industrial PC maker Onlogic and its parent Interlogic for $675 million, the company said on Tuesday. Onlogic keeps its brand and management, and Omron says it wants physical AI services and more US customers.

Perspective Coverage

3 publishers
Builder
Builder 34%
Operator
Operator 23%
Investor
Investor 43%

Reality

Evidence68
Adoption
Not scored
Hype gap+30
Incentives
Not scored
Confidence72
investOne report1 publisher

Ondas will be the first customer of Regna AI, the sovereign model company it wholly owns

Ondas launched Regna AI on October 12, 2026 as a wholly owned company, with Palantir supplying training infrastructure but no equity at the outset. Ondas is also the first customer, so government demand is untested.

Reality

Evidence35
Adoption8
Hype gap+40
Incentives70
Confidence55
buildOne report1 publisher

Deno's move to Cloudflare sets an April 2027 shutdown for Deno Deploy

Deno Deploy shuts down around April 2027 under Ryan Dahl's plan to move the whole Deno team to Cloudflare, a dev.to migration guide reports. Hosted apps must move by then; teams self-hosting Deno get a longer deadline set by the end of security releases.

Publishers:dev.to

Reality

Evidence45
Adoption
Not scored
Hype gap+10
Incentives
Not scored
Confidence40
buildOne report1 publisher

Cloudflare Traces records edge rule, cache and routing decisions as OpenTelemetry spans

Cloudflare has put Traces into open beta, recording each supported step a request takes through its edge as an OpenTelemetry span. The spans export to any OTLP endpoint, so edge decisions can sit in the tracing backend a team already runs.

Reality

Evidence40
Adoption
Not scored
Hype gap+10
Incentives85
Confidence45
buildOne report1 publisher

Cloudflare Workers adds ML-KEM and ML-DSA to Web Crypto behind an opt-in flag

Cloudflare Workers adds five ML-KEM and ML-DSA parameter sets to Web Crypto behind the webcrypto_modern_algorithms flag. Developers can now test post-quantum code against primitives built into the runtime, but they still have to build the protocols on top.

Reality

Evidence60
Adoption
Not scored
Hype gap0
Incentives55
Confidence62
investOne report1 publisher

Companies keep 45% of IT workloads in-house as hyperscalers near half of global capacity

Uptime Institute's survey puts 45% of IT workloads on company-owned infrastructure, unchanged in a year when hyperscalers reached 48% of global capacity. Budgets built on Synergy's forecast of 19% on-premise capacity by 2031 would retire more in-house hardware than that workload split supports.

Publishers:en.sedaily.com

Reality

Evidence45
Adoption40
Hype gap+35
Incentives45
Confidence50
buildConfirmed2 publishers

The Python Workers queue send that needed two imports now takes a plain dict

Cloudflare made Python Workers generally available on September 21, moving binding type conversion into the runtime and adding ASGI and WSGI connectors for FastAPI, Django and Flask. Outbound TCP still goes through JavaScript's connect().

Perspective Coverage

3 publishers
Builder
Builder 70%
Operator
Operator 20%
Investor
Investor 10%

Reality

Evidence60
Adoption
Not scored
Hype gap+20
Incentives65
Confidence70
scienceOne report1 publisher

Microsoft measures a 383% mapping slowdown when robot inference stays onboard

A Microsoft Research study of mobile manipulation reports slower mapping, later obstacle detection and halved manipulation accuracy on small onboard GPUs. The team has shipped Kubernetes tooling to move that inference off the robot.

Publishers:microsoft.com

Reality

Evidence52
Adoption22
Hype gap+24
Incentives78
Confidence46
buildOne report1 publisher

One edge worker in front of hundreds of thousands of tenants makes every deploy all-or-nothing

An InfoQ account of a multi-CDN edge platform describes splitting one Cloudflare Worker into a gateway plus feature workers over service bindings, after the shared script tied every team's cadence to the slowest change in flight.

Publishers:infoq.com

Reality

Evidence35
Adoption25
Hype gap+20
Incentives40
Confidence55
buildOne report1 publisher

Fastly puts a dollar budget and a prompt scan in front of every model call

Fastly's AI Runtime Control is an endpoint you point an application at. It issues virtual keys against your own model provider account, and a key can carry a dollar budget, a rate limit and an ordered list of fallback models.

Publishers:runtimewire.com

Reality

Evidence44
Adoption
Not scored
Hype gap+20
Incentives80
Confidence52
buildOne report1 publisher

SQLite carries the k3s control plane to a claimed 50 nodes and 5,000 pods

A dev.to walkthrough makes the case for k3s over upstream kubeadm on install speed and defaults, and rests it on an embedded SQLite datastore that it rates for 50 nodes and 5,000 pods.

Publishers:dev.to

Reality

Evidence30
Adoption
Not scored
Hype gap+45
Incentives35
Confidence55
leadershipOne report1 publisher

StorMagic's CEO tells retailers to separate the VMware migration from the hardware refresh

Sue Odle argues that a retailer with a thousand stores has to settle one repeatable deployment model before it touches a single site. She runs a company that sells the replacement. Her sequencing argument is testable.

Publishers:forbes.com

Reality

Evidence20
Adoption
Not scored
Hype gap+30
Incentives85
Confidence45
productOne report1 publisher

DataCenterDynamics column moves the UK data centre bottleneck from the grid queue to fibre

The opinion piece argues that securing grid access no longer makes a UK AI site viable, because network access into repurposed industrial plots was never built for AI-scale demand.

Reality

Evidence20
Adoption
Not scored
Hype gap+35
Incentives70
Confidence55
buildOne report1 publisher

Red Hat's two-node OpenShift edge clusters leave one survivor carrying the whole workload

TNA keeps a third etcd vote in a lightweight arbiter; TNF fences the dead node over its BMC. Both keep an edge cluster running on two servers, and both leave one survivor to run everything the site was running.

Publishers:dev.to

Reality

Evidence48
Adoption25
Hype gap+10
Incentives55
Confidence50
productOne report1 publisher

Sandia's analog memory claims 100 times the precision of an unnamed state of the art

Elliot Fuller puts ETCRAM at 100 times the precision of existing analog memory and at least a thousand times the dynamic range. Anyone specifying memory for an edge sensor still has no bit depth and no endurance number.

Reality

Evidence20
Adoption
Not scored
Hype gap+45
Incentives62
Confidence55
buildOne report1 publisher

An 11MB Worker bundle put a three-second compile in front of a 5ms handler

Workers CPU wall time counts only handler execution, so at 0.1 requests per second ScribeToAny's cold isolates spent seconds compiling before the 5ms render, and the platform dashboard showed none of it.

Publishers:dev.to

Reality

Evidence58
Adoption22
Hype gap+18
Incentives62
Confidence55

Earlier coverage