HNHacker News
TopNewBestAskShowJobs

talawahdotnet

509 karma · joined September 3, 2015

submissionscomments
talawahdotnet··on Amazon's cloud business is competing with its own customers
I am personally much happier to see AWS implement managed versions of OSS like Kubernetes and Kafka, than to only see them offer their proprietary versions (Kinesis & ECS).

The fact that all of these services are based on OSS software is what will help keep Amazon in check as they continue to grow and dominate, because at the end of the day if AWS starts to turn into Oracle and get abusive with their pricing/licensing then the customer has a lot more leverage to move to another cloud or self host.

I am not holding my breath for some kind of government driven intervention, nor have I heard an idea of how that would work that makes sense. I think OSS is what is saving us here, and we should celebrate when a project becomes popular enough for AWS to offer it.

Is it fair to the developers that they don't get compensated, nope, but isn't part of the point of OSS to make sure that great solutions get reused and widely deployed?

I think companies building great software can continue to make money servicing it even with AWS in the picture, but I think if you want to make unicorn returns you are going to have to move further up the stack.

Honestly, at some point I hope we are no longer even taking about low level components like Kubernetes and Kafka and that we look back and laugh at the fact that we used to have to license and get support for these things individually. I do also hope that we manage to find a sustainable (if not equitable) model to support the development of these tools.

All the clouds are built on top of the Linux kernel and nobody really expects the kernel devs to get a cut of all those profits, but they seem to have a sustainable model for development, I hope we can continue to figure out good ways to do this for more and more OSS tools.

talawahdotnet··on Amazon's cloud business is competing with its own customers
I really hope not, the fact that all of these services are based on OSS software is what will help keep Amazon in check as they grow, because at the end of the day if AWS starts to turn into Oracle and get abusive with their pricing/licensing then the customer has a lot more leverage to move to another cloud or self host.

To me, the fact that AWS is basing all their services OSS is a good thing for consumers. The thought of it all being proprietary is what would scare me.

talawahdotnet··on Amazon's cloud business is competing with its own customers
I fully agree that AWS should do a better job of contributing back to OSS communities both in terms of code, money and man hours, however I don't think most ppl appreciate that a lot of AWS' secret sauce just doesn't make sense to open source even if they wanted to.

They build their own custom nitro ASICs to integrate with KVM, for networking, storage, security, etc. That code doesn't make sense anywhere but AWS. They build things in ways that are tightly integrated with their internal systems and at a scale that few have experience with.

Take their new Envoy based service mesh for example, I really do hope that they make meaningful contributions back to the main project, but I also suspect they will doing lots of custom integration with their own hardware and internal network that will never make sense to support in an OSS project. Who knows, their Annapurna Labs divsion might even be building a custom card that includes Envoy functionality.

As to them being known for destroying OSS businesses I am drawing a blank there, can you provide some examples?

talawahdotnet··on Amazon's cloud business is competing with its own customers
They don't really need to do anything sneaky to figure out what to build next. They generally build things that their customers are banging down their doors asking them to build.

The main difference with AWS vs Azure or Google is that they have been focused on the space longer and they have really optimized the rate at which they can build out a new scalable, reliable, secure, mananged service.

Sure, the UI/UX is usually not great, and the documentation is imperfect and there are features missing, but they get the hard things right, they charge per second/hour/GB, and they improve over time.

talawahdotnet··on New for AWS Lambda: Use Any Programming Language and Share Common Components
Fargate tasks can now be initiated by CloudWatch Events as well[1]. Lambda will always be at the extreme end of "serverless", but over time I expect the lines to blur more especially since Fargate will now be based on Firecraker[2] which has 125ms startup times for container workloads.

1. https://aws.amazon.com/about-aws/whats-new/2018/08/aws-farga...

2. https://firecracker-microvm.github.io/

talawahdotnet··on New for AWS Lambda: Use Any Programming Language and Share Common Components
I think over time Lambda and Fargate[1] will get closer and closer to each other in terms of functionality and deployment speed.

1. https://aws.amazon.com/blogs/aws/aws-fargate/

talawahdotnet··on Amazon Aurora Global Database
> Amazon Aurora Global Database is a feature that allows a single Amazon Aurora database to span multiple AWS regions. It replicates your data with no impact on database performance, enables fast local reads in each region with typical latency of less than a second, and provides disaster recovery from region-wide outages. In the unlikely event of a regional degradation or outage, a secondary region can be promoted to full read/write capabilities in less than 1 minute.

AWS continues to innovate consistently on Aurora. This is much more approachable and affordable than something like Spanner. Your app definitely needs to be modified to use it, and you need to seriously consider the 1-5s latency impact, but they really do take care of the most difficult parts for you.

My biggest beef is that all the new Aurora features (Serverless, Parallel Query, Global) are only available or MySQL 5.6, not 5.7. God knows how long it will be before they support 8.0

talawahdotnet··on Firecracker – Lightweight Virtualization for Serverless Computing
Sure, there are going to be some people that are excited by the fact that something seems new or just because it is written in Rust, but jessfraz and bcantrill certainly don't fall into those categories. They have a lot of experience with Operating Systems, VMs and containerization and I don't get the impression that they are eaisily impressed by shiny things. Note that they all work for or worked for AWS competitors (Google/MS/Joyent).

I think what is impressive about Firecracker is that they have chosen to reuse a lot of the right things (Linux/KVM/Rust) while also taking a new approach and rethinking important assumptions (No BIOS, no pass-thru, no legacy support, minimal device support).

In my opinion the Firecracker FAQs give sufficient mention to parallel projects and tools they have built on like Kata Containers, QEMU and crosvm. The developers certainly seem open to collaboration with those communities.

AWS doesn't have much of a track record in terms of leading an Open Source projects so some skepticism is understandable, but I think what we have seen so far is a very good start.

talawahdotnet··on Firecracker – Lightweight Virtualization for Serverless Computing
Sure, but if you already have one then there is no incremental cost.

It would be great if they announced that they were gonna remove EKS master costs altogether. Technically Firecracker should make it possible for them to run that infrastructure more efficiently :)

talawahdotnet··on EC2 Instances Powered by Arm-Based AWS Graviton Processors
Yea, I think the fact that it even has a medium instance type is the biggest win for apps that don't need a lot of RAM. c5 and m5 start at large with 4GB/8GB respectively.

To me the most interesting thing will be comparing against t3, which gives you a surprising bang for your buck in scale out workloads, even when you factor in the fractional baseline credit allocation.

talawahdotnet··on EC2 Instances Powered by Arm-Based AWS Graviton Processors
Very interesting. Looking forward to benchmarking these for NodeJS API workloads. Does anybody with experience running node on ARM have any advice/warnings?
talawahdotnet··on Launching EC2 Auto Scaling Groups with mix of Spot and On-Demand Instances
This is awesome! We now have the ability to fully customize the trade off between cost and interruption while still guaranteeing a base level of capacity/reliability.

This should make it much easier to run properly containerized web apps using (a percentage of) Spot instances without worrying.

In some cases the two minute Spot interruption warning time would be sufficient to launch a new instance and avoid a drop in capacity, but I don't get the impression that AWS handles this for you. I need to look into this some more.

Of course you shouldn't ignore the additional complexity this brings, so be sure to understand, design and test the different scaling and failure scenarios!

talawahdotnet··on Faster async functions and promises
Kudos to the work that Google continues to put into optimizing v8. For me async/await is makes a huge difference in terms of the developer experience with Node,js and addressing the remaining performance and debugability issues is a huge win.

The benefits of the v8 engineering efforts continue to multiply beyond just Chrome and Node.js with new uses cases like Cloudflare workers springing up. Even more so when you add WebAssembly to the mix.

talawahdotnet··on OpenCensus – Vendor-agnostic libraries to provide metrics collection and tracing
Man I would really love it if AWS got on board with this (not holding my breath though). After container orchestration, monitoring and tracing is the next major area that I would love to see a standardized API/toolset supported by all Cloud providers.

AWS also needs to revamp Cloudwatch. The UI/UX and agent integrations are severely lacking. Everything feels like it was built by a different team and like updates stopped as soon as they had something functional.

Google's vision with Stackdriver as a single pane of glass to handle logging, errors, metrics and tracing is awesome and as I write this I am seriously considering testing it out even though all my clients run on AWS. Google is clearly positioning Stackdriver for that use case anyway.

talawahdotnet··on Amazon Lightsail Announces 50% Price Drop and Two New Instance Sizes
I wonder if they use their previous generation infrastructure for things like Lightsail. So now that T3s are out they will use the T2 infrastructure for lightsail as customers migrate.
talawahdotnet··on Amazon Aurora Serverless
Yup. I was just saying it is a useful feature.
talawahdotnet··on Amazon Aurora Serverless
It has according to this blog post: https://aws.amazon.com/blogs/aws/aurora-serverless-ga/

"The service currently has autoscaling cooldown periods of 1.5 minutes for scaling up and 5 minutes for scaling down"

talawahdotnet··on Amazon Aurora Serverless
I feel like the choice of the name "serverless" is unnecessarily confusing/controversial, but I like the concept.

The ability to scale down to zero could certainly be useful for automated testing use cases.

As with anything on AWS it will take some testing to discover all the quirks and caveats, but I like the general direction this is headed.

talawahdotnet··on The Node.js Project Introduces Latest Release Line: Node.js 10.x
Yea, built-in is what I was really going for. It would be great if new users could easily do a one-liner fetch without having to:

a) Research and compare the various npm modules available and then install one (which isn't even possible via the Lambda console)

or

b) Figure out how to wrap an event based function in a promise (or search stack overflow and copy one)

Something like this should be super easy to do from the cli/REPL or web console without dependencies. You should only have to install a module when you want more advanced features.

talawahdotnet··on The Node.js Project Introduces Latest Release Line: Node.js 10.x
Good to see that they are adding promisified versions of more core function like fs. It makes things even simpler for new users especially with async/await support.

I really wish they would add a promise based version of the http/https client. The current version uses events rather than the standard callback signature so it can't be automatically converted using util.promisify().

They don't have to modify the existing low-level http client, just add a new API. They could even use the existing fetch() API that users are already familiar with on the client side.

It really should be dead simple for a new user to make an http call in Node without having to npm install a 3rd party library. I recently created an AWS Lambda function (directly via the console) and I just wanted to check to see what the outside world thought my IP was. In theory, with Node.js 8 It should only take a two liner like this:

  const response = await fetch('https://httpbin.org/ip');
  return response.json();
Instead I had to wrap it in a promise myself using untested code that I cribbed from StackOverflow[1].

1.https://gist.github.com/talawahtech/f8c4452dbb543297a008d385...

talawahdotnet··on MySQL 8.0 is now generally available
One of the most interesting new features to me is full support of MySQL 8.0 as a JSON based document store using the new DevAPI[1].

There is a separate blog post highlighting it: https://mysqlserverteam.com/mysql-8-0-announcing-ga-of-the-m...

1. https://dev.mysql.com/doc/x-devapi-userguide/en/devapi-users...

talawahdotnet··on AWS Secrets Manager – Store, Distribute, and Rotate Credentials Securely
The pricing is a bit surprising, but I am sure that if you tried to build the full solution you would end up needing a few more pieces than just the rotation lambda and cron.

Would have been nice if there was a 5 secret free tier though. I feel like the sticker shock will dissuade small users when everybody really should be doing secret rotation.

talawahdotnet··on AWS Secrets Manager – Store, Distribute, and Rotate Credentials Securely
At $0.40 per secret it would have been nice if they had a 5 secret free tier.

That would get smaller users to start using it instead of parameter store and eventually realize the value of automated and audited secret rotation

talawahdotnet··on Keybase is now supported by the Stellar Development Foundation
Are you saying this after having read up on Stellar[1] in particular or is anything other than "full" decentralization a deal breaker for you?

I think Stellar's model if Internet-style (as in the backbone network infrastructure) organic, federated trust relationships is the most practical and viable approach to decentralization that I have seen.

Organizations decide who to trust based on their relationships with them and individuals decide which organizations to trust for the same reason. That is how trust works in the real world, and for good reason. The problem is that the costs of operating at a global level in the current financial system are very high, which leads to centralized control.

Stellar will potentially allow many different types of new, small and innovative organizations to participate in the same network as larger more traditional institutions, with no restrictions for joining the network. Of course those small organization still have to compete for customers in the real world and the playing field is never fully level, but certainly opens up the game.

I personally just don't see the Utopian, fully-decentralized, f*ck-the-system future that a lot of crytpo-enthusiasts envision. It just doesn't mesh with the reality of how non-technical users operate in the real world. For example, right now the vast majority of US users purchase Bitcoin through Coinbase, because that is who they deem trustworthy. What I do see with Stellar is the possibility for greater innovation and greater individual access to the global payments network.

[1] https://medium.com/a-stellar-journey/on-worldwide-consensus-...

talawahdotnet··on Keybase is now supported by the Stellar Development Foundation
They support SQLite as well :)
talawahdotnet··on Keybase is now supported by the Stellar Development Foundation
I could maybe see your concern if Keybase were announcing an ICO for their own token, but that is not what this is.

This really is about Keybase bringing expertise with identity to the table and it makes a lot of sense for the two teams to work together. The currency side of things is still going to be handled by Stellar as far as I can tell.

talawahdotnet··on Amazon MQ – Managed Message Broker Service for ActiveMQ
I believe Apollo was abandoned in favor of Artemis[1] which is based on RedHat's donation of HornetQ. But yea, the community doesn't seem to have rushed to adopt Artemis so it looks like AWS is starting where the numbers are.

The Amazon MQ UI specifically mentions which "Engine" is being used so hopefully there will be other options in the future. Similar to how RDS and Elasticache support multiple DB/Cache Engines.

1. https://activemq.apache.org/artemis/

talawahdotnet··on AWS AppSync – Build data-driven apps with real-time and off-line capabilities
Interesting so the data is actually stored in a datasource like DynamoDB and this is a managed GraphQL service that you can use to handle auth, mapping, subscriptions etc
talawahdotnet··on VPC Endpoints for Your Own Applications and Services
Interesting seems like it will be very useful for providing SaaS offerings to servers that you want to keep extra secure by fully restricting internet access (Private VPC without NAT).

Not 100% clear if it is possible to enable bi-directional access for things like vulnerability scanning, but still seems very handy for a certain target audience.

talawahdotnet··on C5 Instances for EC2 Now Available
Just did a quick test. It booted up in about 11s vs around 19s for Xen.

I did notice that it took a while for the status check to go green though. There was a warning message saying that it couldn't connect to the instance. I was able to SSH just fine though.

← PreviousPage 2 of 3Next →