I keep seeing everyone guessing what the margin is on inference. You say that it's largely break even. We have this person in the thread claim 80% margin (https://news.ycombinator.com/item?id=45462442).
Rate cuts are priced in at about 100% for a 1/4 point. It might go up for a 1/2 point, but that may also signal fear which could have the opposite effect. I think it's hard to say at this point.
The Federal Reserve operates under a "dual mandate" from Congress to promote maximum employment and stable prices (low inflation). As unemployment rises, the Fed will lower interest rates to stimulate investment (increase employment opportunities).
There are definitely some similarities. I think the main difference is the compatibility of Wasm / WASI. Often for unikernel / libOS, you would need to recompile applications to target the specific unikernel / libOS. The goal is that you should be able to take a component that you would run with Wasmtime or other component model compatible runtime and be able to run that on Hyperlight-Wasm.
As for executing a tree of connected components, in the current state of Hyperlight-Wasm you'd probably want to take a collection of components and compose them together using something like https://github.com/bytecodealliance/wac to create a final component composed together from multiple components.
A Wasm component running inside of Wasmtime is just fine. However, when you start to use resources from outside of Wasm, e.g. systems, network interfaces, GPUs, etc., Wasmtime uses OS resources from the host that it is running upon. If this host is running on your trusted compute base, then it implies you are trusting the host implementations in Wasmtime, which for some is just fine. However, Hyperlight-Wasm gives platform builders the ability to describe the interface between the guest and the host explicitly, so you could only expose the host functionality you would want with the trusted implementation you'd want. For example, if I'm building a FaaS, I may want to provide only an exported event handler and an imported key/value interface to the guest for which I've built a safe, multi-tenant implementation and strictly disallow all other host provided functionality.
Krustlet implemented the Kubelet API rather than the containerd-shim API. That means that with containerd shims, you can use the default K8s kubelet, containerd, and extend your containerd runtime with a shim for executing Wasm. Kubelet is much higher in the stack and requires much more to implement it and everything below it in the stack.
I would like to see remote comp be structured with a flat base rate across geos, and then have bonuses for locality to corporate centers. If you feel that it is worth it, you can choose to live close to the office. A flat base rate across geos compensates folks doing the same work for the same pay.
Maybe a good candidate in this organization is one that will leverage every edge they have to produce a desired result. Resourcefulness and risk tolerance are a potent pair.
One of the things we are using it for is for running wasm workloads with cluster networking on K8s. A shim like this makes it simple to construct a small OCI image since all that needs to be included is the wasm module, rather than needing the wasm module and the runtime (500k vs 20MB). Basically, you can define a service and a deployment much the same way as a normal container image and it just works.
What is the common spec for Knative? It seems like one runs an app as normal in a container and then the special scaling sauce is handled by Knative.
When I think of a portable spec for serverless, I think of something more like a Trait of an Interface that needs to be implemented, less a app hosting model. If you think about it like that, then Web Assembly component model [1] would be a great fit for defining an interface that could be implemented in a variety of languages.
Azure has an incubation project which is building a Golang CRD generator and a generic Azure resource controller [k8s-infra]. It will likely end up being used in Azure Service Operator [ASO].
The management cluster is definitely a pattern. Cluster API is designed to have a management cluster to build workload clusters across region and provider.
For quite a few, it is a _main_ goal for the speakers to either draw attention to their company's product or to provide an air of thought leadership on behalf of their company.
The author or company building the software requires money in exchange for using their software. The revenue acquired through the license is then use to pay for additional development.
Revenue sharing seems to imply some kindness / goodwill agreement.
Coming from a rails background, this was the closest out of the box testing experience. Tests are generally handled within a transaction and also offer fixtures and other goodies. Definitely recommend giving it a test drive.
These numbers are complete make believe.