Join us

ContentUpdates from Akatsuki...
Link
@faun shared a link, 1 week, 2 days ago

Serverless RL: Faster, Cheaper and More Flexible RL Training

New product, Serverless RL, available through collaboration between CoreWeave, Weights & Biases, and OpenPipe. Offers fast training, lower costs, and simple model deployment. Saves time with no infra setup, faster feedback loops, and easier entry into RL training...

Link
@faun shared a link, 1 week, 2 days ago

How We Migrated DB 1 to DB 2 , 1 Billion Records Without Downtime

A team movedover 1 billion production records- no downtime, no drama. The stack: dual writes, Kafka retries, and idempotent inserts to keep it clean. They ranshadow readsto sniff for errors, chunked the transfers with checksums, and held off indexing to keep inserts fast. Caches got warmed early to ..

How We Migrated DB 1 to DB 2 , 1 Billion Records Without Downtime
Link
@faun shared a link, 1 week, 2 days ago

Ansible Service Module: Start, Stop, & Manage Services

The Ansibleservicemodulehandles LinuxandWindows without choking on init system quirks. One playbook can start, stop, enable, or restart anything - no matter the OS. Idempotent, so you don’t have to babysit state. Clean and repeatable. Bonus: it’s great for wrangling fleets. Think: coordinating servi..

Link
@faun shared a link, 1 week, 2 days ago

How AWS S3 serves 1 petabyte per second on top of slow HDDs

AWS S3 doesn’t need fancy hardware. It wrings performance out ofcheap HDDs,log-structured merge trees, anderasure coding. The trick? Shard everything. Hit it in parallel. Randomized placementdodges hotspots.Hedged requestsrace the slowest links. And when things get lopsided, S3 rebalances - constant..

How AWS S3 serves 1 petabyte per second on top of slow HDDs
Link
@faun shared a link, 1 week, 2 days ago

Seven Years of Firecracker

AWS is puttingFirecracker microVMsto work in two fresh stacks:AgentCore, the new base layer for AI agents, andAurora DSQL, a serverless, PostgreSQL-compatible database it just rolled out. AgentCore gives each agent session its own microVM. More isolation, less cross-talk - solid for multistep LLM wo..

Seven Years of Firecracker
Link
@faun shared a link, 1 week, 2 days ago

Automated GitHub Self-Hosted Runner Cleanup: Lambda Functions and Auto Scaling Lifecycle Hooks

When an EC2 instance in an Auto Scaling Group shuts down, event-driven plumbing kicks in. Alifecycle hookcatches the scale-in, fires off an SNS notification, and triggers aLambda. That Lambda calls the GitHub API to yank the self-hosted runner before the instance dies. No dangling runners. No manual..

Automated GitHub Self-Hosted Runner Cleanup: Lambda Functions and Auto Scaling Lifecycle Hooks
Link
@faun shared a link, 1 week, 2 days ago

How LogSeam Searches 500 Million Logs per second

LogSeam rips through500M log searches/secand pushes1.5+ TB/s throughputusing Tigris’ geo-distributed object storage. It slashes log volume by 100× with Parquet + Zstandard compression. Then it spins up compute on the fly, right where the data lives—no long-running infrastructure, no laggy reads...

How LogSeam Searches 500 Million Logs per second
Link
@faun shared a link, 1 week, 2 days ago

Internal HTTPS Routing in Istio.

Istio finally bringsinternal HTTPS routingwithSNI-based traffic rules. Services in the mesh can now talk over port 443—TLS fully intact. Just like in prod. TLS terminates at the ingress gateway. Routing pivots on SNI, not headers. Which makes this much closer to real-world mTLS flows. What’s the pla..

Internal HTTPS Routing in Istio.
Link
@faun shared a link, 1 week, 2 days ago

How I Built My Kubernetes Command Toolkit: A Journey from kubectl Chaos to Command Mastery

A dev-built Kubernetes CLI framework reshapeskubectlfor how teams actually work. Commands get grouped by role - dev, SRE, sec, admin - instead of by resource. It bakes in defaults forKyvernopolicies, encourages muscle-memory workflows, and wires up real-time troubleshooting to shrink downtime in pro..

How I Built My Kubernetes Command Toolkit: A Journey from kubectl Chaos to Command Mastery
Link
@faun shared a link, 1 week, 2 days ago

Introducing Headlamp Plugin for Karpenter

The newHeadlamp Karpenter Pluginwires real-time autoscaling insight straight into the Headlamp UI. It showsKarpenterresources, live metrics, scaling moves—no kubectl spelunking required. NodePoolsandNodeClaimsget mapped to core Kubernetes objects. You can tweak configs in the UI, get validation on t..

Introducing Headlamp Plugin for Karpenter

This organization doesn't have a detailed description yet. If you are the administrator of this organization, please claim this page and edit it.