Pertemuan 9: Optimasi Biaya Cloud Progress: 0%

Optimasi Biaya Solusi Cloud

Strategi dan Implementasi Cost Optimization

Financial Focus: Pelajari strategi untuk mengoptimalkan pengeluaran cloud tanpa mengorbankan performa dan keandalan sistem.

Navigasi Cepat

Prinsip Dasar Strategi Optimasi Per Platform Tools & Automation Tugas Praktikum

Cloud Cost Dashboard

Monthly Spend
$2,847.50
Current Month
Potential Savings
$1,243.80
43.7% Reduction
Top Service
EC2
$1,568.30 (55%)
Waste Identified
$892.40
Underutilized Resources

Daftar Materi

Optimization Resources

AWS Pricing Calculator
Cost Estimation Tool
Cost Analysis Template
Excel Spreadsheet
Optimization Scripts
Python & Bash

Tujuan Pembelajaran

Analisis Struktur Biaya

Menganalisis struktur biaya solusi cloud yang telah diimplementasikan

Identifikasi Waste

Mengidentifikasi area pemborosan dalam penggunaan resource cloud

Implementasi Strategi

Mengimplementasikan automated cost optimization di berbagai platform

Prinsip Dasar Cost Optimization

Right-Sizing

Memilih instance type yang tepat untuk workload spesifik

  • Monitor resource utilization
  • Match instance capabilities dengan kebutuhan
  • Avoid over-provisioning
Potential Savings: 40-60%

Elasticity

Memanfaatkan auto-scaling untuk match demand

  • Scale down during off-peak
  • Automatic resource adjustment
  • Pay only for what you use
Potential Savings: 25-40%

Pricing Models

Memilih pricing model yang optimal

  • On-Demand vs Reserved vs Spot
  • Savings Plans commitment
  • Volume discounts
Potential Savings: 50-70%

On-Demand

$0.096
per hour (m5.large)
  • No upfront commitment
  • Maximum flexibility
  • Highest hourly rate
  • Good for unpredictable workloads
RECOMMENDED

Reserved Instances

$0.048
per hour (1-year commitment)
  • 50% savings vs On-Demand
  • 1-3 year commitment
  • Predictable workloads
  • Regional flexibility

Spot Instances

$0.028
per hour (70% savings)
  • Up to 70% savings
  • Can be terminated anytime
  • Fault-tolerant workloads
  • Batch processing

Strategi Optimasi Biaya

Cloud Cost Savings Calculator

High Impact

Right-Sizing Instances

Analyze and match instance types to actual workload requirements

# AWS CLI for instance analysis aws cloudwatch get-metric-statistics \ --namespace AWS/EC2 \ --metric-name CPUUtilization \ --dimensions Name=InstanceId,Value=i-1234567890abcdef0
Savings: 40-60%
High Impact

Spot Instances

Use spot instances for fault-tolerant and flexible workloads

# Launch spot instance aws ec2 request-spot-instances \ --spot-price "0.03" \ --instance-count 1 \ --launch-specification file://specification.json
Savings: 70-90%
Medium Impact

Reserved Instances

Commit to 1-3 year terms for predictable workloads

# Purchase reserved instances aws ec2 purchase-reserved-instances-offering \ --reserved-instances-offering-id "a1b2c3d4" \ --instance-count 5
Savings: 40-60%
Medium Impact

Storage Optimization

Implement lifecycle policies and choose appropriate storage classes

# S3 lifecycle configuration { "Rules": [ { "ID": "Move to Glacier", "Status": "Enabled", "Transition": { "Days": 30, "StorageClass": "GLACIER" } } ] }
Savings: 40-70%

Optimasi per Platform

Cost Comparison Across Platforms

AWS On-Demand
$0.096/hour
AWS Reserved
$0.048/hour
AWS Spot
$0.028/hour
Huawei Cloud
$0.077/hour
OpenNebula
$0.043/hour*
* OpenNebula cost includes infrastructure and operational expenses

OpenNebula Cost Optimization Strategies

Resource Allocation

Dynamic Resource Management

Adjust VM resources based on actual usage patterns

Template Optimization

Use optimized templates to reduce provisioning time and resources

Infrastructure Efficiency

Hardware Consolidation

Consolidate workloads to maximize hardware utilization

Energy Management

Implement power management for non-critical hours

# OpenNebula VM Template for Cost Optimization CPU = "1" MEMORY = "1024" DISK = [ image_id = "5", size = "10000", dev_prefix = "vd" ] SCHED_REQUIREMENTS = "MAX_CPU_USAGE > 50"

AWS Cost Breakdown

EC2 Instances $1,568.30
S3 Storage $234.50
RDS Database $456.80
Data Transfer $89.40
Total Monthly $2,349.00

OpenNebula Cost Breakdown

Hardware Depreciation $1,200.00
Power & Cooling $350.00
Administration $800.00
Software Licensing $150.00
Total Monthly $2,500.00

Tools & Automation

Automated Cost Optimization Scripts

Right-Sizing
Spot Instances
Cost Monitoring
OpenNebula

Automated Right-Sizing Analysis

Spot Instance Automation

#!/bin/bash # spot-instance-manager.sh # Check spot instance pricing INSTANCE_TYPE="m5.large" AZ="us-east-1a" PRICE=$(aws ec2 describe-spot-price-history \ --instance-types $INSTANCE_TYPE \ --availability-zone $AZ \ --product-descriptions "Linux/UNIX" \ --start-time $(date -u +"%Y-%m-%dT%H:%M:%S") \ --end-time $(date -u +"%Y-%m-%dT%H:%M:%S") \ --query 'SpotPriceHistory[0].SpotPrice' \ --output text) echo "Current spot price for $INSTANCE_TYPE in $AZ: $$PRICE" # Launch spot instance if price is favorable MAX_PRICE="0.05" if (( $(echo "$PRICE < $MAX_PRICE" | bc -l) )); then echo "Price is favorable. Launching spot instance..." aws ec2 request-spot-instances \ --spot-price "$MAX_PRICE" \ --instance-count 1 \ --launch-specification file://launch-spec.json fi

OpenNebula Cost Optimization

#!/bin/bash # opennebula-cost-optimizer.sh # Analyze VM utilization and suggest optimizations for VM in $(onevm list --no-header | awk '{print $1}'); do CPU_USAGE=$(onevm show $VM | grep CPU_USAGE | awk -F= '{print $2}') MEMORY_USAGE=$(onevm show $VM | grep MEMORY_USAGE | awk -F= '{print $2}') echo "VM $VM: CPU=$CPU_USAGE%, MEMORY=$MEMORY_USAGE%" # Suggest right-sizing if [ ${CPU_USAGE%.*} -lt 20 ]; then echo " -> Consider downsizing CPU" fi if [ ${MEMORY_USAGE%.*} -lt 25 ]; then echo " -> Consider reducing memory allocation" fi done # Check for idle VMs echo "Checking for idle VMs..." onevm list --no-header | while read line; do VM_ID=$(echo $line | awk '{print $1}') STATUS=$(echo $line | awk '{print $5}') if [ "$STATUS" == "runn" ]; then LAST_ACTIVITY=$(onevm show $VM_ID | grep LAST_POLL | awk -F= '{print $2}') # Logic to detect idle VMs based on last activity echo "VM $VM_ID last activity: $LAST_ACTIVITY" fi done

Tugas Praktikum & Assessment

Tugas Optimasi Biaya Cloud

Tugas 1: Cost Analysis & Right-Sizing

Analisis biaya infrastructure yang sudah diimplementasikan:

  • Identifikasi 3 instances dengan utilization terendah
  • Buat right-sizing recommendations
  • Hitung estimated savings
  • Implementasi perubahan pada 1 instance
Nilai: 40%

Tugas 2: Spot Instances Implementation

Implementasi spot instances untuk workload yang suitable:

  • Analisis spot instance opportunities
  • Implementasi spot fleet
  • Bandingkan cost savings
  • Dokumentasi reliability trade-offs
Nilai: 30%

Tugas 3: Comprehensive Cost Report

Buat laporan optimasi biaya lengkap:

  • Current cost breakdown
  • Identifikasi waste dan inefficiencies
  • Right-sizing recommendations
  • Spot instances opportunities
  • Reserved instances analysis
Nilai: 30%

Deliverables & Timeline

📊
Cost Analysis
Spreadsheet & Charts
⚡
Implementation
Scripts & Configs
📝
Final Report
PDF Document
🎥
Presentation
5-7 minute demo

Kriteria Penilaian

Analysis Accuracy (40%)

  • Correct cost calculations
  • Accurate utilization analysis
  • Valid savings projections
  • Proper waste identification

Implementation (35%)

  • Successful optimizations
  • Script functionality
  • Platform integration
  • Performance maintenance

Documentation (25%)

  • Report completeness
  • Data visualization quality
  • Recommendation clarity
  • Presentation effectiveness

Resources & References

AWS Cost Optimization

OpenNebula Resources

Learning Materials

  • Cloud Financial Management Guide
  • Cost Optimization Best Practices
  • ROI Calculation Templates
  • Case Studies & White Papers