Hacker News Reader

sabalabaon Does the GDPR apply to companies outside of the EU?

Pretty sure that both instances you're talking about (Huawei/Megaupload) are perfect examples of parent's point --- the US is capable of enforcing laws outside of its territory, the EU is not.

sabalabaon Need Something Sorted? Sleep on It

http://chuck.stanford.edu/

Another interesting one is Chuck, a strongly-timed language. :)

Useful for making musical programs.

sabalabaon Ask HN: Who is hiring? (March 2021)

Lambda | Senior Software Engineer | Distributed | Remote | Full Time

Remote first! We're looking for a software engineer who wants to write code at our Deep Learning workstation and server company.

You will:

* Be responsible for the full stack

* Build reliable, scalable, highly available systems

Required Experience:

* Minimum 2 years experience shipping software products

* Bachelor's degree in technical field (CS, Physics, EE, Math, etc.)

* Extremely comfortable in a Linux/Unix shell

===

Lambda is a distributed remote first company headquartered in San Francisco, CA.

Apply to: https://lambdalabs.com/careers

sabalabaon DigitalOcean S-1

> the only two logical outcomes are to either go bust, or to be acquired by a major cloud provider

Or just continue to build a profitable, cash flow positive business that continues to grow for years and years to come. Business outcomes are often not binary.

sabalabaon Nvidia announces mining GPUs, cuts the hash rate of RTX-3060 in half

The CBECI tool estimates that bitcoin consumes about 15.24 gigawatts.

There are approximately 2 million employees working at FDIC insured banks in the United States.[2] There are approximately 330 million people in the United States.[3] United States food production uses about 10 quadrillion BTUs per year (about 332 gigawatts = [10 quadrillion BTU / (365246060) sec to gigawatts]). 332 gigawatts / 330 million ~ 1kW. That's approximately 1 kilowatt per person in the United States. There are probably some good arguments about exports etc but we will hold that aside for now. In 2019, total U.S. primary energy consumption per person (or per capita consumption) was about 305 million British thermal units (Btu).[5] = 305 million btu / (365246060) sec = 10kW. So, we get around 11kW per person for the total food and primary energy consumption rate.

11kW / person * 2 million people = 22 gigawatts

So, we come to the conclusion that if you look at the people employed by FDIC insured banks in the United States alone, they have a larger impact on the environment than bitcoin. If my calculations are correct, maybe we don't need this complaint to be the same first post in every thread on crypto.

Maybe I'm missing something or have done an incorrect calculation, if so, please point it out.

[1] https://www.cbeci.org

[2] https://www.statista.com/statistics/193286/number-of-employe...

[3] https://www.census.gov/popclock/

[4] https://www.chooseenergy.com/blog/energy-101/energy-food-pro...

[5] https://www.eia.gov/tools/faqs/faq.php?id=85&t=1

sabalabaon Ask HN: Who is hiring? (February 2021)

Lambda | Senior Software Engineer | Distributed | Remote | Full Time

Remote first! We're looking for a software engineer who wants to write code at our Deep Learning workstation and server company.

You will:

* Be responsible for the full stack

* Build reliable, scalable, highly available systems

Required Experience:

* Minimum 2 years experience shipping software products

* Bachelor's degree in technical field (CS, Physics, EE, Math, etc.)

* Extremely comfortable in a Linux/Unix shell

===

Lambda is a distributed remote first company headquartered in San Francisco, CA.

Apply to: https://lambdalabs.com/careers

sabalabaon Show HN: Deploy ML Models on a Budget

You should check out Lambda GPU cloud. It's around 40% of the price of AWS/GCP/Azure. The default instance image uses the Lambda Stack repository which is designed for Deep Learning (PyTorch/TensorFlow/Caffe/Drivers pre-installed) saving you time upfront.

If you're still using EC2 you should just switch to us, your monthly bill will thank you.

https://lambdalabs.com/service/gpu-cloud

sabalabaon Show HN: Deploy ML Models on a Budget

If you want to do "budget ML" for training or long-running GPU jobs, you should check out Lambda GPU cloud.

Our on-demand instances are 50% the cost of GCP/AWS. Our 8x V100 instances are $12/hr and our 1x Quadro RTX 6000 instances are $1.25/hr. They come set up with TensorFlow/PyTorch/Jupyter notebook so you don't need to do much DevOps work to get started.

https://lambdalabs.com/service/gpu-cloud

sabalabaon Eliminating nicotine from tobacco products could help millions of smokers quit

Why? The psychoactive effects (relaxation, stimulation, and cognitive enhancement) of Nicotine is the reason that people enjoy tobacco products. Not everybody who smokes tobacco is an addict. Why ruin the fun for everybody else?

While we’re at it, let’s replace all sugar with saccharin, only sell non alcoholic beer, and ban recreational intercourse.

Truly a proposal from a sexless bureaucrat building a teetotalist snoozetown.

sabalabaon A spellchecker used to be a major feat of software engineering (2008)

Interestingly enough, this video from the 1980s shows Brian Kernighan writing a one line spellchecker program in UNIX shell. Obviously, the computer that it’s running on is more powerful than a 256K PC. The point stands: some people are simply living in the future.

https://youtu.be/tc4ROCJYbm0

(Shell coding starts at 8:40)

sabalabaon Why did renewables become so cheap so fast?

It’s not impossible (more like highly likely) for that trust to go the same way of the Nauru Phosphate Royalties Trust. Administrative burden and corruption. Mismanagement. Exponential growth can’t go on forever.

It’s interesting that UAE is doing the same thing that Nauru did except on a larger scale with petrochemicals instead of phosphate.

https://en.wikipedia.org/wiki/Nauru_Phosphate_Royalties_Trus...

sabalabaon Self-Hosting Still Pays

Always love Patrick's posts. Lambda has built a business on moving people from the cloud to on-prem and reducing the cost of cloud AI infrastructure. As some of the other comments have pointed out, for machine learning applications with GPUs, it almost always makes sense to run your workloads on prem, STH is showing that it can also true for traditional CPU workloads.

For example, we're able to provide GPU instances (https://lambdalabs.com/service/gpu-cloud) that are half the hourly cost of AWS hourly on-demand pricing. How? Because there are huge markups on clouds services.

We've done such extensive benchmarking and TCO analysis and the jury is out: it's simply less expensive to run on-prem. You're just paying for convenience when using a cloud. GPU or otherwise.

Sources:

- https://lambdalabs.com/gpu-benchmarks

- https://lambdalabs.com/blog/hyperplane-16-infiniband-cluster...

sabalabaon Ask HN: Who is hiring? (December 2020)

Lambda | Senior Software Engineer | Distributed | Remote | Full Time

Remote first! We're looking for a software engineer who wants to write code at our Deep Learning workstation and server company.

You will:

* Be responsible for the full stack

* Build reliable, scalable, highly available systems

Required Experience:

* Bachelor's degree in technical field (CS, Physics, EE, Math, etc.)

* Extremely comfortable in a Linux/Unix shell

===

Lambda is a distributed remote first company.

Apply to: https://lambdalabs.com/careers

sabalabaon Scaleway Stardust Instances

It looks interesting. Sounds like they're going to be charging for an entire year of service up front or at least as part of an annual contract, which makes sense. (See: "Stardust will only be the beginning in a year-long partnership.") So this will be 21.99 euros ($26.04 USD) for the year. I wonder if this is an x86 instance or if it's ARM. Pricing would make a lot of sense if it was ARM.

sabalabaon Ask HN: Who is hiring? (November 2020)

Lambda | Senior Software Engineer | Distributed | Remote | Full Time

Remote first! We're looking for a software engineer who wants to write code at our Deep Learning workstation and server company.

You will:

* Be responsible for the full stack

* Build reliable, scalable, highly available systems Required Experience:

* Bachelor's degree in technical field (CS, Physics, EE, Math, etc.)

* Extremely comfortable in a Linux/Unix shell

===

Lambda is a distributed remote first company.

Apply to: https://lambdalabs.com/careers

sabalabaon Pandoc – A universal document converter

I absolutely love Pandoc, I use it in my Makefile based static site generator. Pandoc is probably one of the most valuable pieces of open source tooling next to ffmpeg and imagemagick.

sabalabaon Show HN: Lambda Echelon GPU Cluster

Hey HN, just wanted to show you our latest product we just launched at Lambda. Essentially we have an HPC engineer sit down with you, understand your machine learning team's workload, and then design a custom cluster for you.

Here's an hour long tutorial video about how you might roll your own GPU cluster: https://www.youtube.com/watch?v=rfu5FwncZ6s

Here are the slides for the video. http://files.lambdalabs.com/How%20to%20build%20a%20GPU%20clu...

sabalabaon Building a GPU Cluster for AI

Hey HN, this is a talk that I gave at GTC this year. It's a video tutorial on how to build a GPU cluster for AI. It's definitely a bit more enterprise focused, as opposed to what you might do at a startup, but many of the same principles apply.

Here are the slides for the video. http://files.lambdalabs.com/How%20to%20build%20a%20GPU%20clu...

Also, here's the product page for our recently launched GPU cluster product, Echelon: https://lambdalabs.com/gpu-cluster/echelon

sabalabaon Pelican static site generator 4.5

Yea I made a static site generator with about 50 lines of code using make and pandoc:

http://stephenbalaban.com/static-site-generation-in-50-lines...

Most of the work is just creating the index.

sabalabaon Cloudcraft – Architect and budget cloud infrastructure

Trust me, they don't care that you're gone. If you aren't willing to pay $50 / month, you're not their target customer.

sabalabaon Ask HN: Who is hiring? (September 2020)

Lambda | Senior Software Engineer | San Francisco, CA | Remote | Full Time

Remote first! We're looking for a software engineer who wants to write code at our Deep Learning workstation and server company. You will:

* Be responsible for the full stack

* Build reliable, scalable, highly available systems

Required Experience:

* Bachelor's degree in technical field (CS, Physics, EE, Math, etc.)

* Extremely comfortable in a Linux/Unix shell

===

We're located in San Francisco 5 minutes away from the 22nd street Caltrain but the company is a distributed remote first company.

Apply to:

https://lambdalabs.com/careers

sabalabaon Kitaoka’s Popple Illusions – Gabor patches to reverse engineer our visual system

His illusions here were based on a 2000 paper titled "A new illusion demonstrates long-range processing" ( https://www.sciencedirect.com/science/article/pii/S004269890...) That paper showed that there is a global mechanism for orientation averaging. I've always been fascinated with the idea of using illusions to reverse engineer our visual system and this is a really good example of that. The paper is a really accessible read and the rest of Kitaoka's website is a treat to browse:

http://www.psy.ritsumei.ac.jp/~akitaoka/

sabalabaon Adventures in Improving AI Economics

There are 16 characters between the A of Andreessen and the Z in Horowitz for those that don't get it.

sabalabaon Google Earth Timelapse

Some interesting links:

Beijing and Shenzhen's rapid urbanization: https://earthengine.google.com/timelapse#v=39.90419,116.4073...

Deforestation in Brazil: https://earthengine.google.com/timelapse#v=-9.27665,-58.5151...

sabalabaon Building a $5k ML Workstation with Tiitan RTX and Ryzen ThreadRipper [video]

It works for 20.04 LTS and we've updated the website to reflect this, thanks for bringing it to my attention!

sabalabaon Powerful AI Can Now Be Trained on a Single Computer

Yea which are $3 / hour. Or, you could, shameless plug, use Lambda's GPU cloud which has V100s for $1.5 / hour!

https://lambdalabs.com/service/gpu-cloud

I'm embarrassed that I feel so compelled to post this--on a Friday night at that--I apologize.

sabalabaon Building a $5k ML Workstation with Tiitan RTX and Ryzen ThreadRipper [video]

It works for 20.04. Sorry if you didn’t get a reply, I’ll look into it. Where did you email us?

sabalabaon Building a $5k ML Workstation with Tiitan RTX and Ryzen ThreadRipper [video]

No, my comment says that the two 2080 Tis would be better for convnets / situations where you don’t need to train Bert-Large. If you’re sure about memory pooling looking working for DL, please share code and examples, we would love to see one.

sabalabaon TensorFlow, Keras and deep learning, without a PhD

I would disagree that having a lot of interest hurts the field. Where are you seeing “noise and low quality work”? For your sake, I hope you’re not reading random papers from unknown authors on arXiv in your spare time!

I think we’ve seen impressive contribution from people without PhDs. Chris Olah and Alec Radford come to mind first. (Note: I’m not implying that you disagree with that statement, just wanted to point some role models out to those without PhDs who want to contribute to the literature.)

High quality work comes from a tiny incisive fraction of the research community. Most published research, even from PhDs, isn’t worth reading. Easily accessible tutorials promoting Colab are not the problem!

sabalabaon Building a $5k ML Workstation with Tiitan RTX and Ryzen ThreadRipper [video]

Good choice on the 24 GB Titan RTX (so you can do at least batch size = 1 for Bert-Large). Not sure if that's the reason it was chosen though to be honest. If you want to do convnets only then you would do better with NVLinke'd 2080 Tis.

Secondarily, I would suggest that you guys not use windows but instead Ubuntu 18.04 or 20.04 LTS and just install Lambda Stack (https://lambdalabs.com/lambda-stack-deep-learning-software). It's a debian PPA that we maintain at Lambda to keep all of your deep learning drivers, CUDA, CuDNN, TensorFlow, and PyTorch up to date with just apt. It's free!