docs(day-2): New talk

content/day-2/07_pushing-limits.md

@@ -0,0 +1,80 @@
+---
+title: "From us to ms: Pushing Kubernetes Workloads to the Limit"
+weight: 7
+tags:
+ - rejekts
+ - performance
+---
+
+<!-- {{% button href="https://youtu.be/rkteV6Mzjfs" style="warning" icon="video" %}}Watch talk on YouTube{{% /button %}} -->
+
+There were more details in the talk than I copied into these notes.
+Most of them were just too much to write down or application specific.
+
+## Why?
+
+- We need it (Product requirements)
+- Cost efficiency
+
+## Cross Provider Networking
+
+- Throughput:
+    - Same-Zone 200GB/s
+    - Cross-Zone 5-10% Pemnalty
+- Latency:
+    - Same Zone P99: 0.95ms
+    - Cross zone P99: 1.95ms
+- Result: Encourage Services to allways router in the same zone if possible
+- How:
+    - Topology-Aware-Routing (older, a bit buggy)
+    - `trafficDistribution: PreferClose`: Routes to same zone if possible (needs cni-support)
+    - Setup the stack one in each zone
+- Measurements: Kubezonnet can detect cross-zone-traffic
+
+## Disk latency
+
+- Baseline 660MiB/s per SSD aka ~1 SSD per 5GBit/s Networking
+- Example: 100Gbps needs a RAID0 with a bunch of SSDs
+
+```mermaid
+graph LR
+    Querier-->|125ms|Cache
+    Cache-->|200ms|S3
+    direction TB
+    Cache<-->SSD
+```
+
+## Memory managment
+
+- Garbage Collection takes time and is a throughput for latency trade-off
+- Idea: Avoid allocations
+    - Preallocate (e.g. Arenas)
+    - Allocation reuse (e.g. in grpc)
+    - "Allocation Schemes" (thread per core)
+- Avoid memory pressure by
+    - Using gc-friendly types
+    - Tuning your GC
+- Idea: Implement your own optimized data structure
+
+## Optimization in Kubernates
+
+### Defaults
+
+- Best efford
+- No protection from consuming all node memory
+- Critical services could get scheduled on the same node
+
+### Requests and limits
+
+- Requests: Needed to be scheduled
+- Limits: Kill if exceeded
+- Problem: Reactive, it just checks pods according to a cronjob (can be set as apiflag but has a minimum)
+- Downward-API: You can reference the limits in your applications (to let the app trigger gc before the pod gets killed)
+
+### Tains and tolerations
+
+- Pin your workload basted on labels and annotations
+
+### Static cpu manager
+
+- Request a whole number of CPUs -> You get this core guranteed

content/day-2/_index.md

@@ -8,6 +8,9 @@ This is the first day of Cloud Native Rejekts and the first time of me attending
 
 ## Talk recommendations
 
-- Handling large scale migrations: [The Cluster API Migration Retrospective: Live migrating hundreds of clusters to Cluster API](../02_clusterapi) 
+> Ranked by should watch to could watch
-- Handling lrage number of clusters: [CRD Data Architecture for Multi-Cluster Kubernetes](../04_multicluster-crd)
+
 - How to hire, manage and develop engineers: [Tech is broken and AI won't fix it](../05_broken-tech)
+- What if my homelab is an african island: [Geographically Distributed Clusters: Resilient Distributed Compute on the Edge](../06_geo-distributed-clusters)
+- Handling large number of clusters: [CRD Data Architecture for Multi-Cluster Kubernetes](../04_multicluster-crd)
+- Handling large scale migrations: [The Cluster API Migration Retrospective: Live migrating hundreds of clusters to Cluster API](../02_clusterapi) 

content/day-3/_index.md

@@ -8,4 +8,10 @@ Yes that is a negative day.
 Why? Because the numbering of the days is based on KubeCon instead of the trip.
 Why? Ask the sleep deprived version of me who started his trip to London at 2am...
 
-## What h
+## What happened today?
+
+I spent the first seven(-ish) hours of the day travelling from Herzo to London and checked into my hotel.
+After I managed to grab my second breakfeast Tobi texted me that the demo env is down, so I started to prepare an alernative demo.
+
+The afternoon was dedicated to optimizing the demo and getting a screen recording of the demo (harder than it sounds).
+The day ended with a nice thai dinner (and iced thai lemon tea 🤤)