Cost Optimization With Kubernetes
Over the past two years at Magalix, we have focused on building our system, introducing new features, and scaling our infrastructure and microservices. During this time, we had a look at our Kubernetes clusters utilization and found it to be very low. We were paying for resources we didn’t use, so we started a cost-saving practice to increase cluster utilization, use the resources we already had and pay less to run our cluster.
In this article, I will discuss the top five techniques we used to better utilize our Kubernetes clusters on the cloud and eliminate wasted resources, thus saving money. In the end, we were able to cut our monthly bill by more than 50%!
- Applying Workload Right-Sizing
Kubernetes manages and schedules pods are based on container resource specs:
Resource Requests: Kubernetes scheduler is used to place containers on the right node which has enough capacity
Resource Limits: Containers are NOT allowed to use more than their resource limit
Resources requests and limits are container-scooped specs, while multi-container pods define separate resource specs for each container:
apiVersion: apps/v1
kind: Deployment
metadata:
name: nginx
namespace: magalix
spec:
selector:
matchLabels:
app: nginx
replicas: 2
template:
metadata:
labels:
app: nginx
spec:
containers:
- name: nginx
image: nginx:1.14.2
ports:
- containerPort: 80
resources:
requests:
cpu: 100m
memory: 100Mi
limits:
cpu: 1
memory: 1Gi
Kubernetes schedules pods based on resource requests and other restrictions without impairing availability. The scheduler uses CPU and memory resource requests to schedule the workloads in the right nodes, control which pod works on which node and if multiple pods can schedule together on a single node.
Every node type has its own allocatable CPU and memory capacities. Assigning high/unneeded CPU or memory resource requests can end up running underutilized pods on each node, which leads to underutilized nodes.
In this section, we compared resource requests, limited against actual usage and changed the resource request to something closer to the actual utilization while adding a little safety margin.
- Choosing The Right Worker Nodes
Every Kubernetes cluster has its own special workload utilization. Some clusters use memory more than CPU (e.g: database and caching workloads), while others use CPU more than memory (e.g: user-interactive and batch-processing workloads)
Cloud providers such as GCP and AWS offer various node types that you can choose from.
Choosing the wrong node size for your cluster can end up costing you. For instance, choosing high CPU-to-memory ratio nodes for workloads that use memory extensively can starve for memory easily and trigger auto node scale-up, wasting more CPUs that we don’t need.
Calculating the right ratio of CPU-to-memory isn’t easy; you will need to monitor and know your workloads well.
For example, GCP offers general purpose, compute-optimized, memory-optimized with various CPU and memory count and ratios:
Just keep in mind that 1 vCPU is way more expensive than 1GB memory. I have enough memory in the clusters I manage so I try to make sure that when there is a pending pod, this pod is pending on CPUs (which is expensive) so the autoscaler triggers a scale-up for the new node.
To see the cost difference between CPU and memory, let us look at the GCP N2 machine price. GCP gives you the freedom to choose a custom machine type:
(# vCPU x 1vCPU price) + (# GB memory x 1GB memory price)
It’s clear here that the 1vCPU costs 7.44 times more than the cost of 1GB.
To Read More: https://hubs.ly/H0zwcrJ0
Categories
- All Categories
- 217 LFX Mentorship
- 217 LFX Mentorship: Linux Kernel
- 791 Linux Foundation IT Professional Programs
- 353 Cloud Engineer IT Professional Program
- 178 Advanced Cloud Engineer IT Professional Program
- 82 DevOps Engineer IT Professional Program
- 147 Cloud Native Developer IT Professional Program
- 137 Express Training Courses
- 137 Express Courses - Discussion Forum
- 6.2K Training Courses
- 47 LFC110 Class Forum - Discontinued
- 71 LFC131 Class Forum
- 42 LFD102 Class Forum
- 226 LFD103 Class Forum
- 18 LFD110 Class Forum
- 38 LFD121 Class Forum
- 18 LFD133 Class Forum
- 7 LFD134 Class Forum
- 18 LFD137 Class Forum
- 71 LFD201 Class Forum
- 4 LFD210 Class Forum
- 5 LFD210-CN Class Forum
- 2 LFD213 Class Forum - Discontinued
- 128 LFD232 Class Forum - Discontinued
- 2 LFD233 Class Forum
- 4 LFD237 Class Forum
- 24 LFD254 Class Forum
- 697 LFD259 Class Forum
- 111 LFD272 Class Forum
- 4 LFD272-JP クラス フォーラム
- 12 LFD273 Class Forum
- 148 LFS101 Class Forum
- 1 LFS111 Class Forum
- 3 LFS112 Class Forum
- 2 LFS116 Class Forum
- 4 LFS118 Class Forum
- LFS120 Class Forum
- 7 LFS142 Class Forum
- 5 LFS144 Class Forum
- 4 LFS145 Class Forum
- 2 LFS146 Class Forum
- 3 LFS147 Class Forum
- 1 LFS148 Class Forum
- 15 LFS151 Class Forum
- 2 LFS157 Class Forum
- 29 LFS158 Class Forum
- 7 LFS162 Class Forum
- 2 LFS166 Class Forum
- 4 LFS167 Class Forum
- 3 LFS170 Class Forum
- 2 LFS171 Class Forum
- 3 LFS178 Class Forum
- 3 LFS180 Class Forum
- 2 LFS182 Class Forum
- 5 LFS183 Class Forum
- 31 LFS200 Class Forum
- 737 LFS201 Class Forum - Discontinued
- 3 LFS201-JP クラス フォーラム
- 18 LFS203 Class Forum
- 134 LFS207 Class Forum
- 2 LFS207-DE-Klassenforum
- 1 LFS207-JP クラス フォーラム
- 302 LFS211 Class Forum
- 56 LFS216 Class Forum
- 52 LFS241 Class Forum
- 48 LFS242 Class Forum
- 38 LFS243 Class Forum
- 15 LFS244 Class Forum
- 2 LFS245 Class Forum
- LFS246 Class Forum
- 48 LFS250 Class Forum
- 2 LFS250-JP クラス フォーラム
- 1 LFS251 Class Forum
- 152 LFS253 Class Forum
- 1 LFS254 Class Forum
- 1 LFS255 Class Forum
- 7 LFS256 Class Forum
- 1 LFS257 Class Forum
- 1.2K LFS258 Class Forum
- 10 LFS258-JP クラス フォーラム
- 118 LFS260 Class Forum
- 159 LFS261 Class Forum
- 42 LFS262 Class Forum
- 82 LFS263 Class Forum - Discontinued
- 15 LFS264 Class Forum - Discontinued
- 11 LFS266 Class Forum - Discontinued
- 24 LFS267 Class Forum
- 22 LFS268 Class Forum
- 30 LFS269 Class Forum
- LFS270 Class Forum
- 202 LFS272 Class Forum
- 2 LFS272-JP クラス フォーラム
- 1 LFS274 Class Forum
- 4 LFS281 Class Forum
- 9 LFW111 Class Forum
- 259 LFW211 Class Forum
- 181 LFW212 Class Forum
- 13 SKF100 Class Forum
- 1 SKF200 Class Forum
- 1 SKF201 Class Forum
- 795 Hardware
- 199 Drivers
- 68 I/O Devices
- 37 Monitors
- 102 Multimedia
- 174 Networking
- 91 Printers & Scanners
- 85 Storage
- 758 Linux Distributions
- 82 Debian
- 67 Fedora
- 17 Linux Mint
- 13 Mageia
- 23 openSUSE
- 148 Red Hat Enterprise
- 31 Slackware
- 13 SUSE Enterprise
- 353 Ubuntu
- 468 Linux System Administration
- 39 Cloud Computing
- 71 Command Line/Scripting
- Github systems admin projects
- 93 Linux Security
- 78 Network Management
- 102 System Management
- 47 Web Management
- 63 Mobile Computing
- 18 Android
- 33 Development
- 1.2K New to Linux
- 1K Getting Started with Linux
- 371 Off Topic
- 114 Introductions
- 174 Small Talk
- 22 Study Material
- 805 Programming and Development
- 303 Kernel Development
- 484 Software Development
- 1.8K Software
- 261 Applications
- 183 Command Line
- 3 Compiling/Installing
- 987 Games
- 317 Installation
- 97 All In Program
- 97 All In Forum
Upcoming Training
-
August 20, 2018
Kubernetes Administration (LFS458)
-
August 20, 2018
Linux System Administration (LFS301)
-
August 27, 2018
Open Source Virtualization (LFS462)
-
August 27, 2018
Linux Kernel Debugging and Security (LFD440)