6 ms·
Hi all. We're really flattered to be on the home page of hacker news. Thanks!! You caught us by surprise. As you can see we've built in the ability to graceful
by mmaunder 6y ago
Hi all. We're really flattered to be on the home page of hacker news. Thanks!!
You caught us by surprise. As you can see we've built in the ability to gracefully handle a lot of traffic but it's not ideal because you need to wait in a queue until we have enough capacity. So I wanted to jump on here and give you an update of a release we're putting out next week.
FastOrSlow does actual browser simulations from around 12 locations around the world. When we first architected it, we wanted to provide as many locations as possible including places like South Africa. That meant in some cases we needed actual bare-metal servers. So we built it on bare metal in data centers around the world.
We experienced some moderate popularity about a month ago and realized we needed to do a better job of scaling on demand. So we re-evaluated moving on to AWS. Around the same time AWS launched in Cape Town and a few other locations, and so we were able to get the kind of coverage we wanted globally.
So we kicked off an AWS migration. We're days away from moving into AWS and using spot instances to ramp up and down fast based on demand. But as I said, you caught us by surprise.
So please give FastOrSlow a try today, but know that by around late next week we'll be doing a much better job of handling your requests FAST without having to wait in line.
Thanks for your interest and patience. Myself and the lead architect, Ryan, will be here to answer any questions and of course listen to your suggestions.
Regards,
Mark.
- mmaunder 6y agoWe'll also be expanding to 18 locations globally once the migration completes.
- mmaunder 6y agoWe have a lot of jobs in the queue right now. Just passed 1000 folks waiting. Just wanted to reassure you that your job will finish. Just leave the browser window open, take a break and come back. Even if you close your browser, as long as you come back to the same URL, your report will be there once the job is done.
- oyeluckynavin 6y agoIf someone closes the browser, you could remove the user from the queue to save time for now. Lot of users would be closing the browser and might not come back.
- twicetwice 6y agoThis is a really cool tool, and seems much more polished than other similar sites I've used in the past. So first, thanks! Second, though—this seems really expensive to run, based on what you're saying here. How are you monetizing or otherwise funding it?
- mmaunder 6y agoFYI, if you don't want to wait for results (because we're slammed right now via HN) here are a few examples: www.wordfence.com: https://www.fastorslow.com/app/profile/f5fcf22f-8daf-5567-b301-dad0c42cb8ce https://www.fastorslow.com/app/profile/f5fcf22f-8daf-5567-b3... cybersecurity.miami.edu: https://www.fastorslow.com/app/profile/5ad00348-8c5d-566c-8adb-9c795401962d https://www.fastorslow.com/app/profile/5ad00348-8c5d-566c-8a... apple.com: https://www.fastorslow.com/app/profile/a8f2f34f-b695-5ce9-b699-99a0ec4c6989 https://www.fastorslow.com/app/profile/a8f2f34f-b695-5ce9-b6... facebook.com: https://www.fastorslow.com/app/profile/afffcc61-4eb9-5e6c-8840-cd62cea8d84d https://www.fastorslow.com/app/profile/afffcc61-4eb9-5e6c-88... Thanks!! I run Wordfence which does quite well. Profitable and the team is around 35 really great folks with a focus on solid engineering. So we fund it ourselves. For a going concern like us the costs are very modest and we can afford to run it indefinitely without charging. In fact just the bit of brand awareness we get for our main product, Wordfence probably more than justifies the costs. We actually created this as an internal tool because there wasn't anything that would give you a clear indication of site performance from DNS lookup to connect to TTFB to page render from around the world. Our own market is global and even a 1% drop in conversion really costs. After building it we realized that it's very little incremental effort to make it available for all. As I mentioned we're about to migrate into AWS where we run much of our other infrastructure. The financial model and eng design we came up with keeps our costs surprisingly low. We can handle huge traffic spikes and it roughly doubled the cost, even with the spikes. So I'm confident we can run this for the long haul as a free product. A few things I can commit to: Firstly, it's extremely unlikely that you'll see ads on FastOrSlow other than our own company name. That's never been our model and I can't stand when a great site or service gets smothered in ads. In addition, our company has been around since 2006 and we've been made acquisition offers many many times by large companies that you know well. We've consistently turned them down and will continue to do so because we are profitable, love our independence and I think we all know what happens to many great products after acquisition. I'm telling you this because I want you to know that this won't be changing hands to a company that will try to monetize it in obnoxious ways. No plans to offer a paid service right now. Our next release is for internal use and will allow us to do performance surveys across large numbers of sites and present statistically significant insights as research to the online community. Happy to answer any other questions. Thanks for your interest!
- fbelzile 6y agoNice job Mark! The site looks super clean. It's nice to see a post like this explaining what's happening behind the scenes. Remindme! in a week ;)
- mmaunder 6y agoThank you for your kind words. The rest of the team is on this thread and it means a lot to all of us.
- yingw787 6y agoWow, so cool! How does this compare to something like Google Chrome Lighthouse? I think there's a webapp: https://developers.google.com/speed/pagespeed/insights/ https://developers.google.com/speed/pagespeed/insights/ Is there any place we can sign up for updates on when the AWS migration is up and running?
- mmaunder 6y agoI wish we'd thought of that! Once we're in AWS we'll be posting research because we've created an internal tool to do large scale performance surveys e.g. comparing hosting providers or performance of sites using certain libs etc. So I'll put that on the blog and HN might pick it up. FastOrSlow is built using Lighthouse. :) We do actual browser simulations using headless Chrome and lighthouse to get the performance data. We do it programmatically from all the locations, bring the data back and build the report.
- mmaunder 6y agoI did think of one thing: You can follow us on Twitter at: https://twitter.com/isitfastorslow https://twitter.com/isitfastorslow
- battery_cowboy 6y agoWith the positive comments here, perhaps the HN mods wouldn't mind if you posted an update yourself once the AWS migration is completed? I know there's rules around that, but I think they are okay with the occasional self-promotion, especially for great, free tools.
- mmaunder 6y agoWe would be more than happy to chat about how we migrated to AWS, the differences, benefits, down-sides, surprises. Will be a fun post. I'll let rbritton know and we'll put it together, and of course you can try the product too if you like. FastOrSlow has a blog, so we'll post it there.
- nodesocket 6y agoHi, great tool. I am in line (1,201) can you talk a little about how the dynamic spot instances work? Do you kick off spot instance based on CPU load or network usage of all other spot instances in the cluster? Are all regions created equal, or do some regions get more requests than others? I have to imagine your bandwidth costs from AWS aren't gonna be cheap.
- mmaunder 6y agoI think you understand this, but just want to be clear that this is only launching next week. We're currently still on baremetal. It'll be based on sustained queue length averages rather than any kind of load number. We can also bump it up manually. Our estimates for some fairly severe traffic based on historic data are quite reasonable for bandwidth costs. But we're a medium sized business, so we have revenue to support that. I think individuals and very small teams can get really surprised by this. If you're just starting out and don't need dynamic scaling and the tooling that AWS provides, I'd seriously look into colo so that you can buy bandwidth using 95th percentile billing. We use a combination of servers that we own in data centers using 95th billing and AWS. Depends on use case and we have gotten very good at financially modeling future costs for a given config.
- ec109685 6y agoAWS doesn't charge for incoming bandwidth, so the site has advantages there.
- jedberg 6y agoOne thing that might help a bit is to make the profiles you've already done public (well a curated list of popular sites maybe so you don't leak information). Have a list on the home page, so I can see what a report looks like, even if it's not the page I want. Or maybe the page I want will already be there!
- rbritton 6y agoDefinitely worth thinking about. Here's one from yesterday that I also posted below: https://www.fastorslow.com/app/profile/8b282a3c-db98-5b1a-989f-00e88e253f27 https://www.fastorslow.com/app/profile/8b282a3c-db98-5b1a-98...
- mmaunder 6y agoOh hey Ryan. Didn't see you there. :-) Ryan built this thing.
- mmaunder 6y agoOh brilliant. Not sure why I didn't think of this. Here is our company website profile for https://www.defiant.com https://www.defiant.com Profile: https://www.fastorslow.com/app/profile/8b282a3c-db98-5b1a-989f-00e88e253f27 https://www.fastorslow.com/app/profile/8b282a3c-db98-5b1a-98... If you get a 502 or if it takes a few seconds to load, we're busy adding another DB reader, so just reload and it will appear. Should stabilize as the new DB reader comes online. HN is slamming us right now.
- Waterluvian 6y agoHey FWIW I'm not seeing the graceful queue. I'm getting a 429 on my very first request!
- jefftk 6y agoI think that's the queue ;)
- XCSme 6y agoI am 2889 in line, and it doesn't seem to be updating. I have no idea how long it will take, a time estimate would be nice. Is it 1 minute or 1 month?
- rbritton 6y agoIt will depend a bit on how fast or slow the sites are in front, but a general estimate is about 10-15 sites per minute within the limits of the current architecture. Position 2889 will be a while, but it will be today (depending on time zone).
- bobloblaw45 6y agoOh good, I made it look up itself and was worried I broke it for everyone.
- H8crilA 6y agoBe careful, Kurt Goedel once caused a crisis in the fundamentals of mathematics by doing something similar. Also: If you type "Google" into Google - you can break the internet. Nobody has ever tried, not even for a joke.
- JoshTriplett 6y agoWould you consider writing a post on your experiences moving to AWS and spot instances for dynamic scaling? I'd love to hear more about the architectural changes you made for that.
- mmaunder 6y agoAbsolutely. Will do.
- dwheeler 6y agoThis looks like a very cool tool, thank you! I totally understand the "getting surprised". This of all this interest (and the big queue!) as confirmation that this tool is of general interest.
- tozeur 6y agoJust heard the idea for this on a podcast a month ago! On the My First Million podcast
- blep-arsh 6y agoDoes the tool detect when the site responds with a "we don't want you here" page from specific regions? E.g. the Cloudfront blacklist, Cloudflare captcha/proof-of-work challenge, etc. These pages are usually quick to render and may skew results.
- micael_dias 6y agoThey show the screenshot of the page loaded per region so if it did, you'd notice.
- vezycash 6y agoIt seems you aren't caching results. Cos I typed in Apple.com and received a too many checks error instead of just returning a cached result. * A slight tweak to your site allow you to save cash and handle surge traffic with your current server. 1. Returning cached results for sites that's have been recently checked. 2. Collect email addresses and mail the result when it's ready. That's all. It's a much better experience than leaving the tab open, frantically refreshing the page to see if the queue is reducing.