adityapatadia

Not an ad, just a happy user. You can check my profile. I run another startup not affiliated with Mongo.

It’s just me or there are others who keep seeing these updates and think mongodb had all of this years ago?

Seriously so happy to be running our production stack on mongo.

Upvoted for the same reason. We would not be able to run our workloads without Jemalloc. Kudos to this awesome piece of software.

We do. MCP works on top of it.

I say MCP is just lightweight layer on top of REST and as long as it works for end users I don’t see any reason not to adopt it.

This. We also thought MCP was useless, but now the workflows by ChatGPT and Claude are so good; it's as good as installing an app from the App Store. We launched MCP for our SaaS and immediately found users using it.

API and MCP all are just buzzwords for customers. What they care about is whether they can get tasks done in our SaaS using Claude or ChatGPT. If that happens, it's a success.

What a downgrade. We were riding leopards back then!

Building https://www.gumlet.com

Streaming video is still hard to do for a developer today and we are solving that with scalable and cheap infra for streaming.

This change finally helped our company move to typescript.

Just converted many of services to TS and some are WIP. It’s a big win.

We use libvips as well sir. We can't overstate how much we benefit from your work!

We regularly donate to project via open collective. We frankly did not see here due to FB involvement I think.

Jason, here is a story about how much your work impacts us. We run a decently sized company that processes hundreds of millions of images/videos per day. When we first started about 5 years ago, we spent countless hours debugging issues related to memory fragmentation.

One fine day, we discovered Jemalloc and put it in our service, which was causing a lot of memory fragmentation. We did not think that those 2 lines of changes in Dockerfile were going to fix all of our woes, but we were pleasantly surprised. Every single issue went away.

Today, our multi-million dollar revenue company is using your memory allocator on every single service and on every single Dockerfile.

Thank you! From the bottom of our hearts!

For website hosting, it's okay but not great. We encountered issues when we tried to cache a lot of images. Their CDN storage seems really low compared to Cloudflare and Cloudfront. It results in a really bad hit ratio the moment we try to deliver a lot of images.

It’s time saas apps realise that they can’t make 2.5x normal license money by just sticking AI to it.

In our SaaS we added it for free. We realised that there is no way to sustainably make money off of this in long term.

It’s a great feature but not 2.5x price worth feature.

Yes, App Platform. Pricing is a little higher but way lower than AWS but it is fully justified. Zero downtime in the last 1 year.

With Fly, we had 3-4 downtimes in 2023 in a span of 4 months.

We left it about a year ago due to reliability issues. We now use digitalocean apps and working like a charm. Zero downtime with DO.

For image processing, use a proper SaaS dedicated to it: https://www.gumlet.com/image-optimization/

These servers are indeed job processing servers. They are critical but not milliseconds critical. Cooling, security, monitoring, backup generators, and backup of data all are taken care of.

3 years is what we planned. Any extra time we get will be a plus.

we know but it’s actually pretty good.

aditya [at] Gumlet.com

Yes, UPS + 2 independent power lines. Our server has a dual power input PSU. If either power fails, the other one keeps it running.

Nice ideas, but we have chosen a really simple Kubernetes deployment. We only install the host OS (ubuntu server) and then join the self-hosted GPUs as workers in a Kubernetes cluster.

No other task is needed and our Grafana monitors if the server (and its containers) are up and running.

We bought 21. This is just a start.

Tailscale :) We, fortunately, don't need these exposed to the internet so Tailscale works beautifully.

Interestingly, we tried that RTX4000 before we decided to buy our own. Yes, ours will break even in 14 months. ($2300 cost + $40 per month datacenter cost)

I think a single L40 costs $1320 a month on latitude. L40 is also an older GPU.

$40 per server per month. Includes bandwidth, cooling and internet.

Probably yes but we just built it to be cheaper. AM5 is on the costlier side and we don't plan to upgrade these machines. Our calculation is we can retire them by the end of 3 years.

TBH we are at around 15M queries per hour. I am sure our customers don't want us to run on RPi. Btw, it's not only query but billion+ rows which are also there.

May be we are the odd one here but we need that data at millisecond latency (no those are not logs, we use ClickHouse for that)

Just wanted to put here that it's possible to scale Mongo to this level.