HNHacker News
TopNewBestAskShowJobs

thesandlord

1,079 karma · joined October 14, 2014

Cur: Building Mercoa (YC W23)

If you are looking to add Accounts Payable to your B2B product, drop me a line at sandeep at merc○a d○t c○m

Prev: Eng @ Stripe

Founder @ Kbee.app / Simplemnt.com

Developer Advocate @ Google Cloud

https://twitter.com/SandeepDinesh https://medium.com/@SandeepDinesh

[ my public key: https://keybase.io/thesandlord; my proof: https://keybase.io/thesandlord/sigs/f2WVOvjTgL_9f-Z701fGoqiRtlfAavM5RzczPZjZem8 ]

submissionscomments
thesandlord··on Serverless Aurora
The new Firestore database has first-class support for server side SDKs. The real-time database is pretty limited when it comes to server side support, and probably always will be.
thesandlord··on Serverless Aurora
FYI, you can use the Datastore API with Firstore, you just don't get the real-time features. Firestore is backwards compatible with Datastore AFAIK.

(I work for GCP)

thesandlord··on AWS AppSync – Build data-driven apps with real-time and off-line capabilities
Curious to know what limitations you've found. With the new Firestore DB a lot of the old Firebase RTDB scale/query restrictions have been lifted.

(I work for Google)

thesandlord··on EC2 Bare Metal Instances with Direct Access to Hardware
AWS already had dedicated instances, but they still had a VM running on top. These are bare metal, which means you run directly on the hardware.
thesandlord··on Goodbye Google Cloud, Hello Digital Ocean
What you are probably looking for is a PaaS (App Engine, Heroku, Zeit Now, etc) not IaaS

On GCP you can run:

gcloud app deploy

And it will deploy your app on App Engine, with auto-scaling built in. The thing is, that is just the start. Pretty soon you will need databases, caching, versions, etc, etc.

(I work for Google Cloud)

thesandlord··on Tosheets, utility to send stdin to Google sheets
Yeah logger or fluentd are definitely better long term solutions. I wanted something that was no frills and minimal config for ad-hoc situations. Other's on my team found it useful so I open sourced it.

Bonus: In depth code review for this package: https://youtu.be/c5ufcpTGIJM

thesandlord··on Tosheets, utility to send stdin to Google sheets
I wrote something similar that will send stdin to Google Stackdriver Logging.

https://github.com/thesandlord/logpipe

thesandlord··on Azure Functions Now Support Java
GCP employee here.

The Java 8 runtime removed the sandbox restrictions. Most developers hated the restrictions. I highly doubt Microsoft would go down to the path of restrictions given the history in this space.

thesandlord··on Azure Functions Now Support Java
I still don't understand how/why this is serverless. It is FaaS, but not serverless FaaS
thesandlord··on Every Google Cloud Platform product described in 4 words or less
Google Cloud gives you a f1-micro VM for free without a time limit. This about equal to a RPi.
thesandlord··on Linux on Samsung Galaxy smartphones
I had a N810. Great device, but once iPhone and Android smartphones came out, the market for these quickly shrunk to niche status.
thesandlord··on Behind the Masq: Yet More DNS and DHCP Vulnerabilities
> Kubernetes versions 1.5.8, 1.6.11, 1.7.7, and 1.8.0 have been released with a patched DNS pod.
thesandlord··on Wozniak and Copps: Ending net neutrality will end the Internet as we know it
Google: https://cloud.google.com/transfer-appliance/

Microsoft: https://azure.microsoft.com/services/storage/databox/

thesandlord··on Extending per-second billing in Google Cloud
No you are not wrong, and we are working on it. It's going to be a lot more powerful (3.7 GB, 1 full core) than a hobby Dyno, but definitely overkill for testing.

Until then, here is my shameless plug: https://medium.com/google-cloud/three-simple-steps-to-save-c...

thesandlord··on The End of Arduino 101: Intel Leaves Maker Market
> it's the strategic choice to go after the embedded market that I think was unwise

I think the way they entered the market was unwise, not the decision itself. For example, Intel could have created an independent org within itself that could use Intel's resources but wasn't tied to legacy processes.

thesandlord··on The End of Arduino 101: Intel Leaves Maker Market
IBM used to make all of it's money from Mainframes, but then over the years commodity computers have gotten good enough to replace mainframes in most businesses. There are still some customers who need them, but the lion's share of the market is now dominated by commodity x86 servers.

Now x86 is in the process of being disrupted. Look at ARM for example. Currently, most ARM chipsets are focused on mobile applications where low power is necessary. They are not powerful enough to compete with x86 on the server (yes I know there are ARM server parts, but it's still early days), but soon they will be for most customers. And at that point, the power/cost advantage will cause customers to switch over. This is already happening in laptops, with both Apple and Microsoft moving to ARM for their desktop OS. So I think it makes a lot of sense for Intel to try and address this concern, and competing with ARM/MIPS head on for the IoT market is one way to do this.

IMO the mistake Intel made is trying to take their x86 process, which has a value chain designed to serve high margin server markets, and shoehorn it into the IoT market which is dominated by low margin, ultra cheap parts (think ESP8266).

Take a look at this book for more info: https://en.wikipedia.org/wiki/The_Innovator%27s_Dilemma

thesandlord··on Benchmarking TensorFlow on Cloud CPUs: Cheaper Deep Learning Than Cloud GPUs
Compute Engine doesn't really have "Instance Types" or "Instance Families" per say, just Core/Memory combinations. Larger machines have a higher chance of preemption though (according to the PM of PVMs who stated that on this thread).

There are a few interesting projects out there that do the kind of automation you are speaking of like these:

https://github.com/binary-com/gce-manager https://github.com/skelterjohn/prevmtable

Spreading multiple smaller machines over a multi-zone k8s deployment might help mitigate, but it will never solve all the issues.

thesandlord··on Benchmarking TensorFlow on Cloud CPUs: Cheaper Deep Learning Than Cloud GPUs
Google Cloud Developer Advocate here.

Go for it as long as you understand the downside. It's possible that all instances get preempted at once (especially at the 24hr mark), that there isn't capacity to spin up new preemptible nodes in the selected zone once the old instance is deleted, etc. New VMs also take time to boot and join the cluster.

If you are just doing dev/test stuff, I'd recommend using a namespace in your production cluster or spinning up and down test clusters on demand (which can be preemptible).

If you have long running tasks (like a database) or are serving production traffic, using 100% preemptible nodes is not a good idea.

Preemptible can be great for burst traffic and batch jobs, or you can do a mix of preemptible and standard to get the right mix of stability and cost.

thesandlord··on New Singapore Google Cloud Platform region
It's been publicly available for a while (a month maybe?), but was launched quietly until now.
thesandlord··on Kubernetes Production Patterns and Anti-Patterns
Word of caution: Istio isn't production ready yet, still a lot of missing stuff and bugs. Definitely worth playing with though, once it's ready I think it's going to make managing microservices much easier.
thesandlord··on The AWS Spend of Our SaaS Side-Business
Disclaimer: I work for Google Cloud

> Cloud solutions like elastic load balancing are generally engineered for the average use case. Think about the ways you are not like the average.

This might be the case for some things, but many of Google Cloud's "elastic/serverless" products (Load Balancers, App Engine, Pub/Sub, BigQuery, Datastore, etc) can truly scale up from nothing to huge traffic spike without the need for pre-warming and things like that.

thesandlord··on Cloud Spanner is now production-ready
The SQL support is currently for queries only. DML is not supported (yet), so keep that in mind.

(I work for Google Cloud)

Side note: Cosmos is really confusing to me, it has 5 consistency models, supports 4 different database models (I don't think relational is an option?), multiple different APIs, etc.

For example, Azure advertises the p99 latency, but don't specify the consistent model. I'm thinking it's for the eventual consistency, so in that case what's the latency for the strong consistency model? Definitely need to learn more.

thesandlord··on Cloud Spanner is now production-ready
Cassandra is a NoSQL database, so you get horizontal scalability but need to deal with eventual consistency and there is no SQL support. It is similar to Google Bigtable [0].

Spanner is a database that merges the horizontal scalability of a NoSQL database with the strong consistency and SQL semantics of a relational database, basically the best of both worlds. People like to call this type of database "NewSQL."

Cassandra is also open source, Spanner is not. CockroachDB is the OSS clone of Spanner, but there are some differences and tradeoffs to be made as with all things.

(I work at Google Cloud)

[0] https://cloud.google.com/bigtable/

thesandlord··on Google Cloud Prediction API End of Life
Prediction API was closed source and proprietary. Cloud Machine Learning Engine is open source Tensorflow.

You can run these models on prem, on AWS, etc.

thesandlord··on Setting Up HTTP(S) Load Balancing – WebSocket proxy support
You can do URL routing to different backed services like you mentioned, but you can also have a single LB route traffic to the closet datacenter running your service that has capacity. This means you don't need a DNS based system with multiple LBs for a global or HA system.
thesandlord··on Setting Up HTTP(S) Load Balancing – WebSocket proxy support
I'm pretty sure this means Kubernetes Ingress on GCP now support WebSockets as well! People have been asking for this feature for a long time, glad it's available now.

(I work on GCP)

thesandlord··on Going Multi-Cloud with Google Cloud Endpoints and AWS Lambda
https://cloud.google.com/storage/transfer/

Ingress is free on all major clouds, but you still have to pay to get the data out.

(I work on GCP)

thesandlord··on Switch Is Nintendo's 'Fastest Selling' System in History
A laptop will work fine. Even a Android / Win10 tablet with HDMI out will emulate N64/PS1 decently.
thesandlord··on DeepBreath: Preventing angry emails with machine learning
^ waprin is the author of the blog post
thesandlord··on Nintendo Discontinues the NES Classic Edition
It's fairly straightforward, even if you've never set up a RPi before. Once you set up the SD card, it is all GUI based. I'd recommend a RPi3, USB controller(s) (Xbox controllers work great) and a case.

https://retropie.org.uk/

← PreviousPage 7 of 9Next →