aws-lambda-runtimes-performance VS apollo-client

> NodeJs is the slowest runtime, after some time it becomes better(JIT?) but still is not good enough. In addition, we see the NodeJS has the worst maximum duration.

The conclusion drawn about NodeJS performance is flawed due to a quirk of the default settings in the AWS SDK for JS compared to other languages. By default, it opens and closes a TCP connection for each request. That overhead can be greater than the time actually needed to interact with DDB.

I submitted a pull request to fix that configuration[0]. I expect the performance of NodeJS warm starts to look quite a bit better after that.

[0]: https://github.com/Aleksandr-Filichkin/aws-lambda-runtimes-p...

Actually, it's worse than that. Next has started throwing errors if it statically detects you even _importing_ hooks inside of a React Server Component environment:

- https://github.com/apollographql/apollo-client/issues/10974

- https://github.com/apollographql/apollo-client/issues/11167

To the point that Lenz Weber( a maintainer of Apollo Client, and my co-maintainer on Redux Toolkit), is considering resorting to a package that wraps and re-exports all of React's public API just to avoid that static analysis:

- https://github.com/apollographql/apollo-client/pull/11175

Apollo Client (18k ⭐) -> A comprehensive state management library for JavaScript that enables you to manage both local and remote data with GraphQL. Use it to fetch, cache, and modify application data, all while automatically updating your UI.

This currently doesn't work as expected so I have logged this issue in apollo client repo.

I understand this has something to do with the fact, that SvelteKit expects the client and server code to be completely identical - this has already prompted changes to rxjs and apollo is still pending to be fixed. I understand the reason behind being able to run rxjs and apollo on the server but

There is a long-standing issue with error caching (its absence, that is), which leads to errors being rendered as a loading state on the server side. Therefore, if you're doing SSR, you might want to design your schema in such a way that there are no intentional errors in queries whatsoever. To do so, we can use null values instead of NOT_FOUND errors (see how we essentially treated errors as data here?). Note how nothing stops us from setting the response code to 404 in case of a null value, should we want so.

Lol apollo client too.. saw this issue opened like, immediately after the release https://github.com/apollographql/apollo-client/issues/10231

There were some caching issues, outlined here https://github.com/apollographql/apollo-client/issues/6916 to be aware of

If you look at how major backend projects structure their code, it's almost always object-oriented TypeScript.

I submit for the record:

- Apollo Client: https://github.com/apollographql/apollo-client/blob/main/src...

- Storybook: https://github.com/storybookjs/storybook/blob/next/lib/chann...

- Nest: https://github.com/nestjs/nest/blob/master/packages/core/nes...

- MongoDB Driver: https://github.com/mongodb/node-mongodb-native/blob/main/src...

- Prisma: https://github.com/prisma/prisma/blob/main/packages/engine-c...

[※Alert: The following is a code example using the Apollo Client notation, but the Apollo Client does not support Suspense at this time. Github issue about Suspense support: https://github.com/apollographql/apollo-client/issues/9627]