The spectrum of isolation: From bare metal to WebAssembly

The spectrum of isolation: From bare metal to WebAssembly(buildsoftwaresystems.com)

97 points by ThierryBuilds 122 days ago | 34 comments

bmitch3020 122 days ago |

The spectrum comes with multiple tradeoffs, and isn't a simple "bare metal is more secure" narrative. Because as you move into VMs, containers, and code sandboxes, you lose isolation which increases risks, but you also gain capabilities to limit the application which decreases risk. So I believe the most secure approach is layered with much multiple types of isolation working together.

For example, you may isolate a specific customer to bare metal so an escape doesn't compromise other customers. But within that bare metal, you may run containers because they make it easier to work with a read only root filesystem that's also trivial to upgrade. You can also add on user namespaces and seccomp in the container to minimize the risk of a container escape. And then the application may have its own sandbox that limits individual capabilities and which API calls it can run.

Every use case is different, and some layers may not be available depending on that use case. But rather than picking one point on the spectrum, one should pick a list of technologies that best solve each use case.

Loic 121 days ago | |

This is the engineering approach in a factory. You always have multiple layers of security systems.

The analogy is that each layer is a slice of Emmental cheese. You end up with a bad event, only if all the holes in the slices align.

ThierryBuilds 121 days ago | |

Agreed. Beyond security, we must consider other critical factors such as cost, ease of maintenance, and operational overhead.

bflesch 122 days ago |

> This website collects anonymous usage analytics data via GoatCounter and Umami.

My uBlock origin shows that googlefonts.com and fonts.googleapis.com are being blocked.

It irks me a bit that your message explicitly mentions two trackers but it fails to mention the Google tracking. Google is also not mentioned in your privacy policy. Is there a reason for this?

rafram 122 days ago | |

Google Fonts is not a tracker.

https://developers.google.com/fonts/faq/privacy

> For clarity, Google does not use any information collected by Google Fonts to create profiles of end users or for targeted advertising.

bflesch 121 days ago | | |

Google has carte blanche to lie to foreigners for national security purposes, it's not even illegal for them. The data is fed into the mass surveillance systems.

IP, user agent, language headers and network timings are enough to fingerprint and associate you with any other accounts at US tech companies. The visited website is linked via Referer / Origin headers to your browsing history.

All of this tracking is passive and there is no way to check for an independent observer.

Yet here you are defending the most privacy invasive company on the planet.

sillyblob67 121 days ago | | |

By default, loading Google Fonts from Google’s servers exposes user data to Google (e.g., IP Address, User agent, Referrer, Timestamps, Cache identifiers).

It's passive tracking, but it's tracking.

pwdisswordfishy 121 days ago | | |

Well, if Google said it, it must be true.

nake89 122 days ago | |

Your message sent me down a weird rabbit hole of trying to find privacy friendly alternative to google fonts. I found this: https://github.com/coollabsio/fonts They claim to be a privacy friendly drop-in replacement. Their main website: https://fonts.coollabs.io/

cardanome 122 days ago | | |

The easiest solution is to simply self host your fonts.

creata 122 days ago |

venv and sandboxes are such categorically different things that painting it as a spectrum the way this article does is more misleading than helpful.

I also think the article shouldn't mention chroot. From the man page:

> In particular, it is not intended to be used for any kind of security purpose,

I guess it could be part of a sandbox, but there are better tools for that purpose.

(I'm not sure what point there is in giving feedback on an article that's almost entirely LLM-generated, though.)

ThierryBuilds 121 days ago | |

The article is not only about security. Thanks

nickdothutton 122 days ago |

No LPARs (IBM) or LDoms (Oracle), although I appreciate someone might never have to encounter those things these days. They sit above bare metal and below hypervisor VMs.

ThierryBuilds 122 days ago |

I wrote this because I kept seeing developers (myself included) confuse language-level isolation like Python venv with OS-level isolation like Docker. I wanted to trace the actual technical boundaries between them.

The article maps out the differences between common execution environments—from physical bare metal and VMs to containers, process sandboxes, and virtual environments—to create a mental model of where the "isolation boundary" actually sits for each tool.

ianand 122 days ago | |

Since you mention serverless it might be worth mentioning firecracker and v8 isolates.

pjmlp 122 days ago | | |

Or CGIs running on httpd inside HP-UX Vaults, that is how old the idea happens to be.

ThierryBuilds 122 days ago | | |

Thank you for the feedback. I will definitely add them as example solutions for serverless.

fuzzfactor 122 days ago | |

>1. Physical Machine (Bare Metal) This is the foundation.

Nobody should ever forget this.

But I would say this next part is about the opposite for bare metal though:

>Use Case: High-performance computing (HPC), large databases, or legacy systems that require direct hardware access.

To get the utmost reliability out of adequate hardware then bare metal is more suitable for almost everything except for special situations.

Unless something is really wrong with the software or the overall hardware/software approach.

ThierryBuilds 121 days ago | | |

Thanks for the feedback. These are typical use cases where the convenience of higher level abstractions may be less important than the benefits of direct access to the hardware.

lateral_cloud 122 days ago | |

Did you really write it though? Within the first paragraph it's fairly obvious this is heavily LLM-generated.

aragilar 122 days ago | | |

It also has weird definitions. Is nix a virtual environment? Is homebrew a virtual environment? Why is a sandbox different to a container? Type-1 vs Type-2 hypervisors are quite different, and there's no discussion about processes vs threads.

tadfisher 121 days ago | | |

I don't know what it is about LLM-generated text, but when I read it I cannot understand the meaning it is trying to convey. The words are all there, but it is fatiguing to repeatedly parse phrasing like "it's not X but Y" and "you aren't just X, you are Y". The entire article is organized as a sequence of these statements, and this is not hyperbole.

FjordWarden 122 days ago |

Ah, I think I found the reason as to why WebAssembly (in a browser or some other sandboxed environment) is not a suitable substrate for near native performance. It is a very ironic reason: you can't implement a JIT compiler that targets WebAssembly in a sandbox running in WebAssembly. Sounds like an incredibly contrived thing to do but once speed is the goal then a copy-and-patch compiler is a valid strategy for implementing a interpreter or a modern graphics pipeline.

fwsgonzo 121 days ago | |

This is true. A multi-tier JIT-compiler requires writable execute memory and the ability to flush icache. Loading segments dynamically is nice and covers a lot of the ground, but it won't be a magic solution to dynamic languages like JavaScript. Modern WASM emulators already implement a full compiler, linker and JIT-compiler in one, almost starting to look like v8. I'm not sure if adding in-guest JIT support is going in the right direction.

syrusakbary 122 days ago | |

> you can't implement a JIT compiler that targets WebAssembly in a sandbox running in WebAssembly

That's not completely true. With dynamic linking (now supported in WASIX), you can generate and link Wasm modules at runtime easily.

shevy-java 122 days ago |

WebAssembly somehow does not seem to be able to break-through, unlike HTML, CSS, JavaScript did.

mickael-kerjean 122 days ago | |

Or the people who write wasm don't talk too much about it. My OSS work (https://github.com/mickael-kerjean/filestash) has tons of it:

1. to create web versions of applications that are traditionally desktop only to render things like Parquet, PSD, TIFF, SQLite, EPS, ZIP, TGZ, and many more, where C libraries are often the reference implementations. There are almost a hundred supported file formats, most of which are supported through WASM

2. to create plugins that extend the backend and add your own endpoint or middleware as a way to enforce the code run in a constrained environment without the ability to send people's file out

3. in the workflow engine to enable people to run their own sandboxed scripts without giving those a blank check to go crazy

thecupisblue 122 days ago | |

It is more of a silent thing. Running in the background, internal libs, deployment tools, plugin tools.

But also - it's lacking things like a unified positioning + required knowledge to understand it is quite large compared to average dev + most people have no real use for it. It's mostly too "abstract high level" and "low level" for most devs.