Rendered at 13:55:23 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
Normal_gaussian 17 hours ago [-]
I've been using SlicerVM extensively - which is Firecracker MicroVMs for the regular person (and for the irregular with their platform offering) - to run local 'edge' style workloads locally and securly. Agents, local dev CI, etc. It slotted in and replaced my proxmox vm orchestrator, and now I have secure and and fast vms on my laptop wherever I go. It also supports dockerfile style builds if you're wanting a security upgrade from containers (which, you should if you're using agents).
Honestly, while I see firecracker replacing docker on the horizon I don't see firecracker replacing v8 isolates for most edge function execution. Firstly, this article's scenario is a bit unusual in that they were using someone else's isolates - so adding on a few hops; secondly isolates running JS/TS can be statically analyzed quite well, and at scale looking historically for issues and exploits, in many edge compute scenarios this is quite desirable. MicroVMs can have an awful lot more flexibility so to get the same benefit you have to really lock down what is available - the trade-offs for mid-size companies seems to benefit isolates. Obviously netlify is more than big enough and relies heavily on this that it leans in their favour.
dwroberts 6 hours ago [-]
> I see firecracker replacing docker on the horizon
I don’t think this is going to happen because they serve different purposes. Having to boot an entire OS inside a VM is a step backwards compared to containerisation. There are definitely use cases for isolating containers by running them inside a VM (see: Kata) but it generally ‘replacing’ docker, I don’t think that is on the horizon at all
(Also: you need KVM available, you need a rootfs to boot which will be larger than a container, it has no built in support for mounts or really any kind of communication with the host unless you explicitly set up networking etc for it)
jeroenhd 5 hours ago [-]
> Having to boot an entire OS inside a VM
The whole point of Firecracker over normal VM solutions is that "booting an OS" takes milliseconds.
Still, I don't think Docker is at risk of being replaced just yet because of the resource management benefits you get from sharing a kernel.
Normal_gaussian 3 hours ago [-]
Balloon devices go a long way for ram sharing, though not all the way. CPU scheduling is great.
There's definitely a tradeoff, but IME it's not as drastic as you first think.
dwroberts 3 hours ago [-]
Whether it takes milliseconds or seconds though, you are still doing it. You’re still just running a VM
troupo 2 hours ago [-]
> The whole point of Firecracker over normal VM solutions is that "booting an OS" takes milliseconds.
Define "OS". A barebones OS will indeed start in microseconds. But then your app/service won't. Because it will likely need a gazillion things that a barebones OS doesn't provide. Suddenly the startup time isn't microseconds or even milliseconds.
Sadly, we never got the promise of unikernels, and everything requires the full-blown OS to run anything
jeroenhd 35 minutes ago [-]
The VM's I'm talking about don't run some kind of weird bespoke OS, just a normal Linux kernel. You can make Linux boot quite quickly if you strip out everything but the hardware you virtualise, and if you resume a snapshot from CoW RAM, you don't even need to do that.
I don't have any recent measurements myself, but the startup time for a full Linux kernel using Firecracker's snapshotting feature seems to be around 30ms if you optimise calls for latency: https://dev.to/adwitiya/how-i-built-sandboxes-that-boot-in-2...
The measurements in that blog post place Firecracker above Docker in terms of startup performance. Most of that delay is probably the Docker daemon rather than Linux namespace APIs, but that's beside the point.
Smolvm with it's libkrun vmm provides significantly worse security positioning than slicervms use of firecracker, which leads to slicervm for any dangerous or secure workload.
Outside of that I support GPU and enables something called branchable computing.
dprkh 7 hours ago [-]
Why is this better than just using Docker?
binsquare 6 hours ago [-]
Kernel level isolation.
Functionality of criu built in so you can get rewind, pause, in an accessible manner.
Embeddable (you can write JavaScript to programmatically use an isolated environment)
Native performance on multiplatform + consistent experience across platforms.
stavros 5 hours ago [-]
This sounds great. By the way, does "kernel-level isolation" mean "you have to allocate a chunk of RAM to this"? Or is that CPU-level?
EDIT: Ah, looks like it means "runs its own kernel", not "isolates at the kernel" like Docker does.
binsquare 4 hours ago [-]
yes, separate kernels + virtualized hardware via hypervisor.
containers are built on linux primitives & so shares the kernel.
stavros 3 hours ago [-]
Excellent, thank you. This is definitely useful to me.
QGQBGdeZREunxLe 16 hours ago [-]
I think Alex learned his lesson offering a free version running OpenFaaS.
alexellisuk 5 hours ago [-]
The world: "build a secure, enterprise-ready microVM automation solution - work on it full time, and pay salaries for the staff that work on it"
Also: it has to be free.
So yes you're right, people confuse VC backed companies, and vibe-coded pet-projects for sustainable software.
SlicerVM was started in 2022 and internal only, plenty of YouTube videos and such about it - written completely manually from our Actuated work.
There's a free trial for anyone who wants to play about on their Mac or Linux computer, the comment here is from a real user (unprompted) that knew and used free alternatives previously.
tietjens 6 hours ago [-]
Would love to read a blog post about this if you've every time to cobble one together.
If someone in Netlify is listening, could you please add support for Fetchable[1] in Netlify Edge Functions? I opened a ticket 1 week ago but it has gone unanswered[2].
The idea behind Fetchable is to have a (semi) standard across runtimes and environments to be able to do this:
Even Node.js, who are traditionally the laggards in these things, are already working on it and implementing it. This would be very useful for library writers like me to be able to standardize code across envrionments. I'm re-writing npm's `server` (currently WIP here [3]) and I have some ugly workarounds ONLY for Netlify Edge Functions.
I guess I am curious if/how exactly the Netlify announcement relates to unikernels?
yencabulator 15 hours ago [-]
> In the past, requests went out to a hosted execution service. Today, they run on MicroVMs inside our own edge network — roughly 5x faster at the median.
So, the execution itself might now be slower as far as we know, they just eliminated some networking from the mix? Misleading.
nchmy 18 hours ago [-]
I'm having trouble understanding/believing this, given that Cloudflare Workers are also v8 isolates and run vastly faster than the 25-40ms that netlify says their isolates took...
phickey 18 hours ago [-]
from the article: "With our old infrastructure it went out over the internet, ran the edge function, and came back to us to pass on. With the new compute platform, the request is forwarded to a compute node within our network."
As far as I know, Cloudflare Workers have always executed within Cloudflare's network, not gone out to the internet and executed elsewhere (which I read as being in a hyperscaler cloud).
nchmy 6 hours ago [-]
OK, so it's less about isoates vs firecracker than it is about them now doing it in-house...
Perhaps the article says that, but the framing is all wrong
irq-1 18 hours ago [-]
> In the past, requests went out to a hosted execution service. Today, they run on MicroVMs inside our own edge network
The isolates were not being run at the edge.
bobfunk 17 hours ago [-]
They were running on the edge, and in the same datacenters but by another provider.
ascorbic 8 hours ago [-]
That provider being Deno
jedberg 17 hours ago [-]
The next time you want to curse AWS, remember they gave us Firecracker, one of the best microvm technologies out there, and the basis of at least a few non-AWS products out there (this one being the newest entry to the list).
Onavo 11 hours ago [-]
I don't think anybody hates AWS as a technology. It's mostly the rent-seeking (pricing almost-free bandwidth to an egregious level, nickel and diming users through fractional pricing calculations that are about as transparent as the US health insurance billing, predatory B2B vendor lock-in practices) that gets on people's nerves.
When MBA (and YC startup) schools teach to build product with a "high switching cost", they do not have consumer's (in this case the developers) interests in mind.
The other issue is that outside of the core offerings (S3, EC2/Lambdas, the logging and queuing services), everything else seems to be in a perpetual state of beta with products shipped by interns. The situation is not as bad as Cloudflare but the bar's on the ground. Pre-LLM, services like Cognito caused developers no end of pain and suffering. Poor documentation, buggy products opaque console UIs, there's no end to complaints when it comes to AWS. If their services were designed well, then companies like Vercel and Heroku shouldn't exist at all.
And what's worse, with all of their efforts at squeezing customers, they still pay the worst of out of all of the big techs for non-senior leadership day to day engineering ICs. At least with Facebook there's commensurate pay. It's like Amazon took a look at Asian "996" culture and figured that if it works for their warehouse staff it should be good for the engineering folks too.
gottorf 10 hours ago [-]
> companies like Vercel and Heroku shouldn't exist at all
Crazy that Heroku with its head start has become abandonware. Goes to show that no incumbent is immune to change.
fragmede 9 hours ago [-]
Blame Salesforce.
WatchDog 9 hours ago [-]
All else being equal, v8 isolates should be lower latency than firecracker.
As they mention, the v8 latency was because the v8 runtime was hosted externally.
Firecracker has better security model.
This old thread[0] with kentonv author of Cloudflare's workers platform covers some of the differences.
This is highly interesting considering AWS invented the MicroVMs for lambda, yet node on lambda is dog slow (both in latency and throughput). I can traumadump on request. I bet they could use some of this tech especially since their use cases are often not too dissimilar (auth validation, rule checking etc)
notatoad 13 hours ago [-]
seriously - if netlify can do 15ms warm start + 10ms cold start, wtf is lambda doing for the other 1500ms?
tomnipotent 16 hours ago [-]
> they could use some of this tech
Firecracker is AWS tech, and was behind both Lambda & Fargate. I imagine it's all the enterprisey extras built on top that cause those issues.
nderjung 16 hours ago [-]
It is originally AWS, though we forked Firecracker ~4 years ago now and maintain an internal version with custom hypercalls (command-and-control from within the VM, POSIX-style fork, etc.) and a number of other features. We regularly pull patches and fixes from upstream Firecracker. You can read about the differences here:
> Over the past several months, our team has rebuilt the infrastructure behind Edge Functions, working closely with the team at Unikraft, who wrote about the experience from their side.
nderjung 12 hours ago [-]
Netlify use Unikraft for microVMs; just explaining the variant of Firecracker they are using.
fragmede 14 hours ago [-]
When would you prefer they advertise? Seems entirely relevant.
tomnipotent 13 hours ago [-]
When they're contributing to the conversation. It's not relevant.
fragmede 13 hours ago [-]
How is it not relevant? In what way doesn't it contribute?
tomnipotent 13 hours ago [-]
It's on you to prove it does. In what ways does it contribute?
fragmede 13 hours ago [-]
We now know about their unikraft VM, whereas previous we did not.
tomnipotent 13 hours ago [-]
If only there was an HN guideline about not using the site primarily for promotion.
pjmlp 5 hours ago [-]
As fly by comment, I bet very few HNers actually read those guidelines.
fragmede 12 hours ago [-]
You said they didn't contribute and weren't relevant. Whether their account primarily exists for promotion is a separate question.
ascorbic 8 hours ago [-]
It's relevant because it's about TFA! I'm assuming you didn't read it: Netlify uses Unikraft MicroVMs
zokier 15 hours ago [-]
fargate is not firecracker
binsquare 15 hours ago [-]
can confirm that this is true for the fargate product used by consumers.
"Firecracker was developed at Amazon Web Services to accelerate the speed and efficiency of services like AWS Lambda and AWS Fargate."
CodesInChaos 16 hours ago [-]
> When the MicroVM boots up and the JavaScript server begins to listen on a port, we take a snapshot of the MicroVM. [...] we start a new MicroVM from that snapshot.
That sounds scary, since forked RNG states can lead to catastrophic failures in UUID generators or cryptography.
You’d hope so, though it’s not unheard of to get this wrong. Fastly managed to snapshot guest seeds and use them at runtime for RNG in their WASM snapshots a few years ago https://nvd.nist.gov/vuln/detail/cve-2022-39218
14 hours ago [-]
sorenbs 7 hours ago [-]
We are running Postgres and Bun on the same tech at Prisma. It’s good.
aaronvg 19 hours ago [-]
Wish it explained where the v8 isolate latency is coming from compared to microvms
wmf 18 hours ago [-]
"In the past, requests went out to a hosted execution service."
They were outsourcing to another company so there's plenty of room for overhead to creep in.
vmg12 18 hours ago [-]
v8 isolates aren't actually a great sandbox and I would not trust them implicitly in the AI era. This is probably why they wrap them in an additional sandbox.
torginus 17 hours ago [-]
But I guess they are good enough to isolate multiple instances of the same code, ran by the same customer in parallel.
Normal_gaussian 17 hours ago [-]
Without commenting on v8 isolates specifically, this doesn't necessarily hold in any isolation situation; many customers are running code on behalf of their customers, which are often submitting jobs on behalf of theirs, and so on. Isolation breaches within a platform customer can result in significant cross-user data breaches.
torginus 6 hours ago [-]
You're right, but there are cases when the risk can be managed. Like if you're running an auth lambda in one isolate, and another isolate is running the exact same copy of the code, I'd say malicious exploitation would be low enough a risk, that I'd be comfortable running things like this.
And I'd say this even is a majority use case.
dummydummy1234 18 hours ago [-]
Why are v8 isolates bad, I see speculative execution hacks, but are there others?
phickey 18 hours ago [-]
Despite naming them isolates, the V8 team does not consider them to be a security boundary.
binsquare 18 hours ago [-]
v8 isolates are still shared kernel
while microvm's are separate kernel + hardware virtualization through hypervisor guarantees
I wouldn't call it bad either, just different tools for different things
londons_explore 17 hours ago [-]
Not only are they shared kernel... They're shared process, shared address space, shared memory pool and allocator... In fact, there is very little isolated about them at all.
I bet there are a million ways to cause side channels allowing learning about other code or data on the same machine, and just one V8 bug (of which there have historically been thousands) let's you take over or modify code in another isolate.
vmg12 17 hours ago [-]
The v8 JIT is very complex and can lead to sandbox escapes if there are type confusion bugs.
pjmlp 5 hours ago [-]
They would be even faster with AOT compiled code.
groundzeros2015 11 hours ago [-]
I don't think isolates are good idea. We need kernel level process isolation. Chrome itself doesn't trust isolates.
ur-whale 7 hours ago [-]
> 5x faster Edge Functions: V8 isolates to Firecracker MicroVMs
Off-topic, but ...
I've noticed over the last 5 years I find myself increasingly incapable of understanding the title of HN posts upon first reading them.
And then sometimes also, reading the article itself does not help either.
The likely explanation is that I'm probably getting old and senile, but maybe, just maybe, I am not the only one experiencing this.
This post's title is a decent example, albeit far from the worse I've read.
I think HN should have a feature to graph the level of incomprehensibility of post titles over time.
An LLM should probably be able to measure this automatically.
masklinn 2 hours ago [-]
> The likely explanation is that I'm probably getting old and senile, but maybe, just maybe, I am not the only one experiencing this.
Here it looks like you haven’t gotten interested or kept up with some of the field, none of “edge functions”, “v8 isolates”, and “microvms” / “firecracker microvms” are ultra novel or made up, just relatively specialised.
tinykit 8 hours ago [-]
[flagged]
ContinuityLab 13 hours ago [-]
[dead]
TaupeRanger 19 hours ago [-]
[flagged]
CharlesW 19 hours ago [-]
This is technical content, written for a technical audience, being shared on HN. In that context, it's not particularly esoteric. If "5x faster Edge Functions" means nothing to you, that's a good signal that you can just skip it without FOMO. If you're familiar with the concept but not Firecracker (a popular microVM) specifically, the expectation is that you'll use search engines and/or LLMs to get up to speed.
secondcoming 18 hours ago [-]
If you're counting milliseconds why use Javascript?
wmf 18 hours ago [-]
V8 is extremely optimized for script startup time.
pjmlp 5 hours ago [-]
So the whole rewrite node scripts into Rust and Go isn't really needed. /s
pjmlp 5 hours ago [-]
The whole rewrite into compiled languages hasn't yet reached serverless circles.
Honestly, while I see firecracker replacing docker on the horizon I don't see firecracker replacing v8 isolates for most edge function execution. Firstly, this article's scenario is a bit unusual in that they were using someone else's isolates - so adding on a few hops; secondly isolates running JS/TS can be statically analyzed quite well, and at scale looking historically for issues and exploits, in many edge compute scenarios this is quite desirable. MicroVMs can have an awful lot more flexibility so to get the same benefit you have to really lock down what is available - the trade-offs for mid-size companies seems to benefit isolates. Obviously netlify is more than big enough and relies heavily on this that it leans in their favour.
I don’t think this is going to happen because they serve different purposes. Having to boot an entire OS inside a VM is a step backwards compared to containerisation. There are definitely use cases for isolating containers by running them inside a VM (see: Kata) but it generally ‘replacing’ docker, I don’t think that is on the horizon at all
(Also: you need KVM available, you need a rootfs to boot which will be larger than a container, it has no built in support for mounts or really any kind of communication with the host unless you explicitly set up networking etc for it)
The whole point of Firecracker over normal VM solutions is that "booting an OS" takes milliseconds.
Still, I don't think Docker is at risk of being replaced just yet because of the resource management benefits you get from sharing a kernel.
There's definitely a tradeoff, but IME it's not as drastic as you first think.
Define "OS". A barebones OS will indeed start in microseconds. But then your app/service won't. Because it will likely need a gazillion things that a barebones OS doesn't provide. Suddenly the startup time isn't microseconds or even milliseconds.
Sadly, we never got the promise of unikernels, and everything requires the full-blown OS to run anything
I don't have any recent measurements myself, but the startup time for a full Linux kernel using Firecracker's snapshotting feature seems to be around 30ms if you optimise calls for latency: https://dev.to/adwitiya/how-i-built-sandboxes-that-boot-in-2...
The measurements in that blog post place Firecracker above Docker in terms of startup performance. Most of that delay is probably the Docker daemon rather than Linux namespace APIs, but that's beside the point.
Huggingface places Firecracker startup time between 100-300ms while Docker is at 50-200ms, based on Kata: https://huggingface.co/blog/agentbox-master/firecracker-vs-d...
Either way, latency is in the same ballpark of "low enough that it only matters in edge cases".
Fwiw, you can run this instead free and open source: https://github.com/smol-machines/smolvm
Disclaimer: Am author.
https://github.com/libkrun/libkrun
Firecracker has a long track record but has a lot of knobs and tunings to get the security right.
smolvm's serve mode confines each VMM by default with a seccomp allowlist, Landlock, a per-VM uid and no_new_privs, much like Firecracker's jailer.
For dangerous workloads, people can do the same things such as skip host mounts and use virtio-net.
It's not a different security class just because it's libkrun vs firecracker
any point of comparison with microsandbox? [0]
[0] https://github.com/superradcompany/microsandbox
Good sandboxing is a feature of a good VM.
Outside of that I support GPU and enables something called branchable computing.
Functionality of criu built in so you can get rewind, pause, in an accessible manner.
Embeddable (you can write JavaScript to programmatically use an isolated environment)
Native performance on multiplatform + consistent experience across platforms.
EDIT: Ah, looks like it means "runs its own kernel", not "isolates at the kernel" like Docker does.
containers are built on linux primitives & so shares the kernel.
Also: it has to be free.
So yes you're right, people confuse VC backed companies, and vibe-coded pet-projects for sustainable software.
SlicerVM was started in 2022 and internal only, plenty of YouTube videos and such about it - written completely manually from our Actuated work.
There's a free trial for anyone who wants to play about on their Mac or Linux computer, the comment here is from a real user (unprompted) that knew and used free alternatives previously.
The idea behind Fetchable is to have a (semi) standard across runtimes and environments to be able to do this:
Even Node.js, who are traditionally the laggards in these things, are already working on it and implementing it. This would be very useful for library writers like me to be able to standardize code across envrionments. I'm re-writing npm's `server` (currently WIP here [3]) and I have some ugly workarounds ONLY for Netlify Edge Functions.[1] https://fetchable.org/ [2] https://answers.netlify.com/t/support-fetch-for-netlify-edge... [3] https://server-js.com/
We also did a couple of technical write ups if you're interested:
- https://unikraft.com/blog/netlify-edge-functions
- https://unikraft.com/customer-stories/edge-functions-netlify
I guess I am curious if/how exactly the Netlify announcement relates to unikernels?
So, the execution itself might now be slower as far as we know, they just eliminated some networking from the mix? Misleading.
As far as I know, Cloudflare Workers have always executed within Cloudflare's network, not gone out to the internet and executed elsewhere (which I read as being in a hyperscaler cloud).
Perhaps the article says that, but the framing is all wrong
The isolates were not being run at the edge.
When MBA (and YC startup) schools teach to build product with a "high switching cost", they do not have consumer's (in this case the developers) interests in mind.
The other issue is that outside of the core offerings (S3, EC2/Lambdas, the logging and queuing services), everything else seems to be in a perpetual state of beta with products shipped by interns. The situation is not as bad as Cloudflare but the bar's on the ground. Pre-LLM, services like Cognito caused developers no end of pain and suffering. Poor documentation, buggy products opaque console UIs, there's no end to complaints when it comes to AWS. If their services were designed well, then companies like Vercel and Heroku shouldn't exist at all.
And what's worse, with all of their efforts at squeezing customers, they still pay the worst of out of all of the big techs for non-senior leadership day to day engineering ICs. At least with Facebook there's commensurate pay. It's like Amazon took a look at Asian "996" culture and figured that if it works for their warehouse staff it should be good for the engineering folks too.
Crazy that Heroku with its head start has become abandonware. Goes to show that no incumbent is immune to change.
As they mention, the v8 latency was because the v8 runtime was hosted externally.
Firecracker has better security model.
This old thread[0] with kentonv author of Cloudflare's workers platform covers some of the differences.
[0]: https://news.ycombinator.com/item?id=31740885
Firecracker is AWS tech, and was behind both Lambda & Fargate. I imagine it's all the enterprisey extras built on top that cause those issues.
https://unikraft.com/blog/unikraft-vs-firecracker
> Over the past several months, our team has rebuilt the infrastructure behind Edge Functions, working closely with the team at Unikraft, who wrote about the experience from their side.
as someone who worked on it
https://github.com/firecracker-microvm/firecracker
"Firecracker was developed at Amazon Web Services to accelerate the speed and efficiency of services like AWS Lambda and AWS Fargate."
That sounds scary, since forked RNG states can lead to catastrophic failures in UUID generators or cryptography.
They were outsourcing to another company so there's plenty of room for overhead to creep in.
And I'd say this even is a majority use case.
while microvm's are separate kernel + hardware virtualization through hypervisor guarantees
I wouldn't call it bad either, just different tools for different things
I bet there are a million ways to cause side channels allowing learning about other code or data on the same machine, and just one V8 bug (of which there have historically been thousands) let's you take over or modify code in another isolate.
Off-topic, but ...
I've noticed over the last 5 years I find myself increasingly incapable of understanding the title of HN posts upon first reading them.
And then sometimes also, reading the article itself does not help either.
The likely explanation is that I'm probably getting old and senile, but maybe, just maybe, I am not the only one experiencing this.
This post's title is a decent example, albeit far from the worse I've read.
I think HN should have a feature to graph the level of incomprehensibility of post titles over time.
An LLM should probably be able to measure this automatically.
Here it looks like you haven’t gotten interested or kept up with some of the field, none of “edge functions”, “v8 isolates”, and “microvms” / “firecracker microvms” are ultra novel or made up, just relatively specialised.