lettering your resources autoscale uncontrollably, that's a bill issue
September 9, 2023 at 2:28 AM
lettering your resources autoscale uncontrollably, that's a bill issue
I am kicking ass this week!
Yesterday, got the Gateway API/LoadBalancer configured.
Today, got the backend microservices configured to autoscale based on demand.
Tomorrow, finish what I started with data persistence.
Friday, configure the nodes themselves to autoscale.
Yesterday, got the Gateway API/LoadBalancer configured.
Today, got the backend microservices configured to autoscale based on demand.
Tomorrow, finish what I started with data persistence.
Friday, configure the nodes themselves to autoscale.
April 3, 2025 at 12:59 AM
I am kicking ass this week!
Yesterday, got the Gateway API/LoadBalancer configured.
Today, got the backend microservices configured to autoscale based on demand.
Tomorrow, finish what I started with data persistence.
Friday, configure the nodes themselves to autoscale.
Yesterday, got the Gateway API/LoadBalancer configured.
Today, got the backend microservices configured to autoscale based on demand.
Tomorrow, finish what I started with data persistence.
Friday, configure the nodes themselves to autoscale.
Some fun Judoscale stats to end the year!
🚀 Over 17 million autoscale events
😮 Largest single autoscale event was upscaling 80 dynos at once!
📈 64 billion web requests
🧑🤝🧑 We now help over 1,000 teams keep their apps up and healthy
Enjoy our full year-in-review here: judoscale.com/blog/judosc...
🚀 Over 17 million autoscale events
😮 Largest single autoscale event was upscaling 80 dynos at once!
📈 64 billion web requests
🧑🤝🧑 We now help over 1,000 teams keep their apps up and healthy
Enjoy our full year-in-review here: judoscale.com/blog/judosc...
Judoscale’s 2024 Year-End Review
judoscale.com
December 23, 2024 at 3:15 PM
Some fun Judoscale stats to end the year!
🚀 Over 17 million autoscale events
😮 Largest single autoscale event was upscaling 80 dynos at once!
📈 64 billion web requests
🧑🤝🧑 We now help over 1,000 teams keep their apps up and healthy
Enjoy our full year-in-review here: judoscale.com/blog/judosc...
🚀 Over 17 million autoscale events
😮 Largest single autoscale event was upscaling 80 dynos at once!
📈 64 billion web requests
🧑🤝🧑 We now help over 1,000 teams keep their apps up and healthy
Enjoy our full year-in-review here: judoscale.com/blog/judosc...
📦 cboxdk/laravel-queue-autoscale v4.3.3
Intelligent, predictive autoscaling for Laravel queues with SLA/SLO-based optimization
🔗 https://github.com/cboxdk/laravel-queue-autoscale
Intelligent, predictive autoscaling for Laravel queues with SLA/SLO-based optimization
🔗 https://github.com/cboxdk/laravel-queue-autoscale
September 24, 2026 at 1:59 PM
📦 cboxdk/laravel-queue-autoscale v4.3.3
Intelligent, predictive autoscaling for Laravel queues with SLA/SLO-based optimization
🔗 https://github.com/cboxdk/laravel-queue-autoscale
Intelligent, predictive autoscaling for Laravel queues with SLA/SLO-based optimization
🔗 https://github.com/cboxdk/laravel-queue-autoscale
no longer will this be an issue! it's now built on a cluster that can autoscale (still on-prem), but won't have to for a very long time because of how ridiculously efficient it is
here's cpu usage on the old cluster vs. the new
here's cpu usage on the old cluster vs. the new
July 10, 2026 at 10:54 AM
no longer will this be an issue! it's now built on a cluster that can autoscale (still on-prem), but won't have to for a very long time because of how ridiculously efficient it is
here's cpu usage on the old cluster vs. the new
here's cpu usage on the old cluster vs. the new
some vanilla mashups you just see and go 'yep thats for me'
quick and dirty model copy/autoscale test? trying it out since @wardeneater.bsky.social has been doing it a bunch (but fem to masc), doesn't look too bad
> www.xivmodarchive.com/modid/127765
quick and dirty model copy/autoscale test? trying it out since @wardeneater.bsky.social has been doing it a bunch (but fem to masc), doesn't look too bad
> www.xivmodarchive.com/modid/127765
January 22, 2025 at 3:21 PM
some vanilla mashups you just see and go 'yep thats for me'
quick and dirty model copy/autoscale test? trying it out since @wardeneater.bsky.social has been doing it a bunch (but fem to masc), doesn't look too bad
> www.xivmodarchive.com/modid/127765
quick and dirty model copy/autoscale test? trying it out since @wardeneater.bsky.social has been doing it a bunch (but fem to masc), doesn't look too bad
> www.xivmodarchive.com/modid/127765
It's not like AWS, you can't just autoscale if you got the ratio wrong
October 29, 2025 at 3:30 AM
It's not like AWS, you can't just autoscale if you got the ratio wrong
viva o autoscale nas infras em nuvem mas com pena do time de dev ainda assim
Bluesky activity is now measured by brazilians' hours of sleep
September 5, 2024 at 3:07 PM
viva o autoscale nas infras em nuvem mas com pena do time de dev ainda assim
Does That include the autoscale Update?
April 1, 2025 at 6:56 AM
Does That include the autoscale Update?
This week on the Learn Kubernetes Weekly:
🍯 Securing Kubernetes with honeypots
💻 Autoscale with My MacBook Screen Lock
💣 Ingress production incident
🙈 YAML Like a String
✅ JobSet and GKE metrics
⭐️ Densify
Read it now: https://learnkube.com/issues/139
🍯 Securing Kubernetes with honeypots
💻 Autoscale with My MacBook Screen Lock
💣 Ingress production incident
🙈 YAML Like a String
✅ JobSet and GKE metrics
⭐️ Densify
Read it now: https://learnkube.com/issues/139
July 9, 2025 at 11:31 AM
This week on the Learn Kubernetes Weekly:
🍯 Securing Kubernetes with honeypots
💻 Autoscale with My MacBook Screen Lock
💣 Ingress production incident
🙈 YAML Like a String
✅ JobSet and GKE metrics
⭐️ Densify
Read it now: https://learnkube.com/issues/139
🍯 Securing Kubernetes with honeypots
💻 Autoscale with My MacBook Screen Lock
💣 Ingress production incident
🙈 YAML Like a String
✅ JobSet and GKE metrics
⭐️ Densify
Read it now: https://learnkube.com/issues/139
Autoscaling #HuggingFace Text Generation Interface (TGI) on #GKE. We just published a video with @wietsevenema.eu.
In this video we demonstrate how to autoscale TGI workloads on Google Kubernetes Engine (GKE) using TGI queue size as the scaling signal.
www.youtube.com/watch?v=QjLZ...
In this video we demonstrate how to autoscale TGI workloads on Google Kubernetes Engine (GKE) using TGI queue size as the scaling signal.
www.youtube.com/watch?v=QjLZ...
How to autoscale a TGI deployment on GKE
YouTube video by Google Cloud Tech
www.youtube.com
November 26, 2024 at 5:32 PM
Autoscaling #HuggingFace Text Generation Interface (TGI) on #GKE. We just published a video with @wietsevenema.eu.
In this video we demonstrate how to autoscale TGI workloads on Google Kubernetes Engine (GKE) using TGI queue size as the scaling signal.
www.youtube.com/watch?v=QjLZ...
In this video we demonstrate how to autoscale TGI workloads on Google Kubernetes Engine (GKE) using TGI queue size as the scaling signal.
www.youtube.com/watch?v=QjLZ...
Docker + VM but as soon as you want to autoscale, you should go to more complex system like SaaS Kubernetes.
February 18, 2025 at 10:28 AM
Docker + VM but as soon as you want to autoscale, you should go to more complex system like SaaS Kubernetes.
Un énorme merci à OVH (poke @aurelievache.bsky.social ^^) pour l'infra !!
Version anglaise : a-cup-of.coffee/blog/karpent...
Version anglaise : a-cup-of.coffee/blog/karpent...
Karpenter + Cluster API
My journey to autoscale an OVH cluster deployed via the Cluster API, using Karpenter and its cluster-api provider, with MKS as the management cluster.
a-cup-of.coffee
July 12, 2026 at 9:55 PM
Un énorme merci à OVH (poke @aurelievache.bsky.social ^^) pour l'infra !!
Version anglaise : a-cup-of.coffee/blog/karpent...
Version anglaise : a-cup-of.coffee/blog/karpent...
אז ניתקתי מגע ואני לא הולך לריב איתם. אבל בכל מקום שבו יש אנשים יהיו אינטריגות. פעוטות ככל שיהיו, אבל הן יהיו שם.
ואנשים יעשו מהן משהו משמעותי בגלל הautoscale, ואז יתחילו הפיצוצים.
ואנשים יעשו מהן משהו משמעותי בגלל הautoscale, ואז יתחילו הפיצוצים.
September 1, 2025 at 2:07 PM
אז ניתקתי מגע ואני לא הולך לריב איתם. אבל בכל מקום שבו יש אנשים יהיו אינטריגות. פעוטות ככל שיהיו, אבל הן יהיו שם.
ואנשים יעשו מהן משהו משמעותי בגלל הautoscale, ואז יתחילו הפיצוצים.
ואנשים יעשו מהן משהו משמעותי בגלל הautoscale, ואז יתחילו הפיצוצים.
Check out @stebet.net's blog post on how we rolled our own autoscale for @haveibeenpwned.com using Azure's Hyperscale SQL Server (and the things we learned along the way about how costs are calculated 😲)
Azure SQL HyperScale is awesome and works wonders for
@haveibeenpwned.com. But it isn't perfect. We ran into some issues with how scaling in Azure SQL HyperScale works, so I decided to write a blog post about it.
stebet.net/autoscaling-...
@haveibeenpwned.com. But it isn't perfect. We ran into some issues with how scaling in Azure SQL HyperScale works, so I decided to write a blog post about it.
stebet.net/autoscaling-...
Autoscaling Azure SQL HyperScale for better cost management
Learn how Have I Been Pwned optimizes Azure SQL HyperScale cloud costs by dynamically scaling vCores and memory usage through automated scripts.
stebet.net
March 20, 2025 at 1:17 AM
Check out @stebet.net's blog post on how we rolled our own autoscale for @haveibeenpwned.com using Azure's Hyperscale SQL Server (and the things we learned along the way about how costs are calculated 😲)
sì, anyway oramai le autoscale, i c.d. "cestelli", sono alla portata di qualsiasi colletta istantanea
March 31, 2025 at 12:15 PM
sì, anyway oramai le autoscale, i c.d. "cestelli", sono alla portata di qualsiasi colletta istantanea
espero que não tenha sido no autoscale infinito. provável que tenha sido no meu cartão 👹
September 7, 2024 at 11:20 PM
espero que não tenha sido no autoscale infinito. provável que tenha sido no meu cartão 👹
if it's well-engineered, it's a couple scaling changes (or traffic change + autoscale)
which i would hope it is, but i have no direct contact with ant's engineering culture
and it's also a fast growing company, which can lead to more corner-cutting in engineering
which i would hope it is, but i have no direct contact with ant's engineering culture
and it's also a fast growing company, which can lead to more corner-cutting in engineering
June 15, 2026 at 3:48 PM
if it's well-engineered, it's a couple scaling changes (or traffic change + autoscale)
which i would hope it is, but i have no direct contact with ant's engineering culture
and it's also a fast growing company, which can lead to more corner-cutting in engineering
which i would hope it is, but i have no direct contact with ant's engineering culture
and it's also a fast growing company, which can lead to more corner-cutting in engineering
you can add 's', 'm', or 'l' after the filename and before the .jpg or .png in an imgur URL, and it will autoscale the images down for you.
If you’re over 30, quote this with some life advice 🤌🏼
May 10, 2025 at 4:36 AM
you can add 's', 'm', or 'l' after the filename and before the .jpg or .png in an imgur URL, and it will autoscale the images down for you.
I'm one of those people who defends its value, but I also think it can use improvements. I still think spell slot damage should autoscale without heightening, as the issues that it caused in 1e aren't present in 2e, and I do believe the spell rank spread gets a bit obtuse to manage.
April 28, 2025 at 1:11 AM
I'm one of those people who defends its value, but I also think it can use improvements. I still think spell slot damage should autoscale without heightening, as the issues that it caused in 1e aren't present in 2e, and I do believe the spell rank spread gets a bit obtuse to manage.
"One queue per worker" is the way. Sidekiq, Solid Queue, doesn't matter.
In other words, each queue should have a dedicated worker process.
But isn't that wasteful??
Nope. You autoscale them down to zero when they’re idle. They’ll start back up when jobs are enqueued.
In other words, each queue should have a dedicated worker process.
But isn't that wasteful??
Nope. You autoscale them down to zero when they’re idle. They’ll start back up when jobs are enqueued.
April 8, 2025 at 4:46 PM
"One queue per worker" is the way. Sidekiq, Solid Queue, doesn't matter.
In other words, each queue should have a dedicated worker process.
But isn't that wasteful??
Nope. You autoscale them down to zero when they’re idle. They’ll start back up when jobs are enqueued.
In other words, each queue should have a dedicated worker process.
But isn't that wasteful??
Nope. You autoscale them down to zero when they’re idle. They’ll start back up when jobs are enqueued.
אני חושב שכן. אבל אני חושב שבכל מקום שבו לא תהיי יש autoscale של הגרף. וקיצוניים יהיו בכל מקום, ותמיד יהיו אנשים שימצאו על מה לריב עם אנשים אחרים.
אני רואה את זה ביישוב שלי, עם אנשים שראיתי בהם מענטשים, ושממש התאכזבתי מההתייחסות שלהם בנקודות בזמן.
אני רואה את זה ביישוב שלי, עם אנשים שראיתי בהם מענטשים, ושממש התאכזבתי מההתייחסות שלהם בנקודות בזמן.
September 1, 2025 at 2:07 PM
אני חושב שכן. אבל אני חושב שבכל מקום שבו לא תהיי יש autoscale של הגרף. וקיצוניים יהיו בכל מקום, ותמיד יהיו אנשים שימצאו על מה לריב עם אנשים אחרים.
אני רואה את זה ביישוב שלי, עם אנשים שראיתי בהם מענטשים, ושממש התאכזבתי מההתייחסות שלהם בנקודות בזמן.
אני רואה את זה ביישוב שלי, עם אנשים שראיתי בהם מענטשים, ושממש התאכזבתי מההתייחסות שלהם בנקודות בזמן.
I'm pretty sure Blue Sky's NOC said "fuck it" and clicked autoscale in AWS.
November 15, 2024 at 1:24 PM
I'm pretty sure Blue Sky's NOC said "fuck it" and clicked autoscale in AWS.