Gateway Performance

Overview

This section provides detailed performance specifications and benchmarks for all Aviatrix gateway types. Performance data includes throughput, connection limits, latency characteristics, and resource utilization across different deployment scenarios.

Transit Gateway Performance

High-Performance Transit Gateways

AWS Performance
Azure Performance
GCP Performance

C5n Instance Performance

Instance Type	vCPUs	Memory	Network Performance	Max Throughput	Connection Limit
c5n.large	2	5.25 GB	Up to 25 Gbps	1.5 Gbps	10,000
c5n.xlarge	4	10.5 GB	Up to 25 Gbps	3 Gbps	25,000
c5n.2xlarge	8	21 GB	Up to 25 Gbps	6 Gbps	50,000
c5n.4xlarge	16	42 GB	25 Gbps	12 Gbps	100,000
c5n.9xlarge	36	96 GB	50 Gbps	25 Gbps	250,000
c5n.18xlarge	72	192 GB	100 Gbps	50 Gbps	500,000

C6i Instance Performance

Instance Type	vCPUs	Memory	Network Performance	Max Throughput	Connection Limit
c6i.large	2	4 GB	Up to 12.5 Gbps	1.2 Gbps	8,000
c6i.xlarge	4	8 GB	Up to 12.5 Gbps	2.5 Gbps	20,000
c6i.2xlarge	8	16 GB	Up to 12.5 Gbps	5 Gbps	40,000
c6i.4xlarge	16	32 GB	12.5 Gbps	10 Gbps	80,000
c6i.8xlarge	32	64 GB	25 Gbps	20 Gbps	160,000
c6i.16xlarge	64	128 GB	50 Gbps	40 Gbps	320,000

D-Series Performance

VM Size	vCPUs	Memory	Network Performance	Max Throughput	Connection Limit
D2s_v3	2	8 GB	Moderate	1 Gbps	8,000
D4s_v3	4	16 GB	Moderate	2 Gbps	16,000
D8s_v3	8	32 GB	High	4 Gbps	32,000
D16s_v3	16	64 GB	High	8 Gbps	64,000
D32s_v3	32	128 GB	Very High	16 Gbps	128,000
D64s_v3	64	256 GB	Extremely High	30 Gbps	256,000

F-Series Performance

VM Size	vCPUs	Memory	Network Performance	Max Throughput	Connection Limit
F2s_v2	2	4 GB	Moderate	1.2 Gbps	10,000
F4s_v2	4	8 GB	Moderate	2.5 Gbps	20,000
F8s_v2	8	16 GB	High	5 Gbps	40,000
F16s_v2	16	32 GB	High	10 Gbps	80,000
F32s_v2	32	64 GB	Very High	20 Gbps	160,000
F64s_v2	64	128 GB	Extremely High	35 Gbps	320,000

N2 Instance Performance

Machine Type	vCPUs	Memory	Network Performance	Max Throughput	Connection Limit
n2-standard-2	2	8 GB	Up to 10 Gbps	1 Gbps	8,000
n2-standard-4	4	16 GB	Up to 10 Gbps	2 Gbps	16,000
n2-standard-8	8	32 GB	Up to 16 Gbps	4 Gbps	32,000
n2-standard-16	16	64 GB	Up to 32 Gbps	8 Gbps	64,000
n2-standard-32	32	128 GB	Up to 32 Gbps	16 Gbps	128,000
n2-standard-64	64	256 GB	Up to 50 Gbps	32 Gbps	256,000

C2 Instance Performance

Machine Type	vCPUs	Memory	Network Performance	Max Throughput	Connection Limit
c2-standard-4	4	16 GB	Up to 10 Gbps	2.5 Gbps	20,000
c2-standard-8	8	32 GB	Up to 16 Gbps	5 Gbps	40,000
c2-standard-16	16	64 GB	Up to 32 Gbps	10 Gbps	80,000
c2-standard-30	30	120 GB	Up to 32 Gbps	20 Gbps	160,000
c2-standard-60	60	240 GB	Up to 50 Gbps	40 Gbps	320,000

Spoke Gateway Performance

Application-Optimized Performance

General Purpose
High Performance

Instance Type	Platform	vCPUs	Memory	Max Throughput	Connection Limit	Use Case
t3.medium	AWS	2	4 GB	500 Mbps	5,000	Development
t3.large	AWS	2	8 GB	750 Mbps	7,500	Small Apps
t3.xlarge	AWS	4	16 GB	1.5 Gbps	15,000	Medium Apps
B2s	Azure	2	4 GB	480 Mbps	4,800	Development
B2ms	Azure	2	8 GB	720 Mbps	7,200	Small Apps
B4ms	Azure	4	16 GB	1.4 Gbps	14,000	Medium Apps
e2-medium	GCP	1	4 GB	400 Mbps	4,000	Development
e2-standard-2	GCP	2	8 GB	600 Mbps	6,000	Small Apps
e2-standard-4	GCP	4	16 GB	1.2 Gbps	12,000	Medium Apps

Instance Type	Platform	vCPUs	Memory	Max Throughput	Connection Limit	Use Case
c5.large	AWS	2	4 GB	1 Gbps	10,000	Web Services
c5.xlarge	AWS	4	8 GB	2 Gbps	20,000	API Services
c5.2xlarge	AWS	8	16 GB	4 Gbps	40,000	Database Apps
F2s_v2	Azure	2	4 GB	1.2 Gbps	12,000	Web Services
F4s_v2	Azure	4	8 GB	2.4 Gbps	24,000	API Services
F8s_v2	Azure	8	16 GB	4.8 Gbps	48,000	Database Apps
c2-standard-4	GCP	4	16 GB	2.5 Gbps	25,000	Web Services
c2-standard-8	GCP	8	32 GB	5 Gbps	50,000	API Services
c2-standard-16	GCP	16	64 GB	10 Gbps	100,000	Database Apps

Edge Gateway Performance

VPN and Security Performance

UserVPN Performance

Gateway Size	Platform	Concurrent Users	Throughput per User	Total Throughput	SSL VPN Connections
Small	All	100	50 Mbps	1 Gbps	100
Medium	All	250	40 Mbps	2 Gbps	250
Large	All	500	30 Mbps	4 Gbps	500
XLarge	All	1,000	25 Mbps	8 Gbps	1,000
XXLarge	All	2,000	20 Mbps	15 Gbps	2,000

Site-to-Site VPN Performance

Gateway Size	IPSec Tunnels	Throughput per Tunnel	Total Throughput	BGP Sessions
Small	10	100 Mbps	1 Gbps	10
Medium	25	80 Mbps	2 Gbps	25
Large	50	80 Mbps	4 Gbps	50
XLarge	100	80 Mbps	8 Gbps	100
XXLarge	200	75 Mbps	15 Gbps	200

Firewall Performance

Gateway Size	Firewall Rules	Connections/sec	Concurrent Connections	Inspection Throughput
Small	1,000	5,000	50,000	800 Mbps
Medium	2,500	12,500	125,000	1.6 Gbps
Large	5,000	25,000	250,000	3.2 Gbps
XLarge	10,000	50,000	500,000	6.4 Gbps
XXLarge	20,000	100,000	1,000,000	12 Gbps

Performance Optimization Guidelines

Instance Sizing Recommendations

Transit Gateway Sizing

Small Deployments (< 1 Gbps)

AWS: c5n.large or c6i.large
Azure: D2s_v3 or F2s_v2
GCP: n2-standard-2 or c2-standard-4

Medium Deployments (1-10 Gbps)

AWS: c5n.2xlarge or c6i.4xlarge
Azure: D8s_v3 or F8s_v2
GCP: n2-standard-8 or c2-standard-8

Large Deployments (> 10 Gbps)

AWS: c5n.9xlarge or c6i.16xlarge
Azure: D32s_v3 or F32s_v2
GCP: n2-standard-32 or c2-standard-30

Spoke Gateway Sizing

Development/Test

Low traffic: t3.medium/B2s/e2-medium
Moderate traffic: t3.large/B2ms/e2-standard-2

Production Applications

Web services: c5.large/F2s_v2/c2-standard-4
API services: c5.xlarge/F4s_v2/c2-standard-8
Database applications: c5.2xlarge/F8s_v2/c2-standard-16

Edge Gateway Sizing

VPN Workloads

Small: < 100 users or 10 site connections
Medium: 100-250 users or 25 site connections
Large: 500+ users or 50+ site connections

Security Workloads

Small: < 1,000 firewall rules
Medium: 1,000-5,000 firewall rules
Large: 5,000+ firewall rules

Performance Monitoring

Key Performance Indicators

Throughput utilization vs capacity
Connection count vs limits
CPU and memory utilization
Network latency and packet loss

Scaling Triggers

70% throughput utilization
80% connection capacity
75% CPU utilization sustained
Latency exceeding SLA thresholds

Performance Tuning Tips

Network Optimization
Security Optimization
Monitoring & Tuning

Bandwidth Optimization

Enable enhanced networking on supported instances
Use placement groups for high-bandwidth applications
Optimize MTU sizes for your network environment
Configure appropriate buffer sizes for high-throughput workloads

Latency Optimization

Deploy gateways in regions closest to workloads
Use dedicated tenancy for consistent performance
Optimize routing to minimize network hops
Configure TCP optimization settings for long-distance connections

Performance specifications are based on laboratory testing under optimal conditions. Real-world performance may vary based on network conditions, traffic patterns, security policies, and other environmental factors.

Best Practice: Start with smaller instance sizes and scale up based on actual performance measurements rather than theoretical requirements. Monitor key performance indicators and scale proactively to maintain optimal performance.

Concepts & Architecture

Guides

Reference

Overview

Transit Gateway Performance

High-Performance Transit Gateways

Spoke Gateway Performance

Application-Optimized Performance

Edge Gateway Performance

VPN and Security Performance

Performance Optimization Guidelines

Instance Sizing Recommendations

Transit Gateway Sizing

Spoke Gateway Sizing

Edge Gateway Sizing

Performance Monitoring

Performance Tuning Tips

Concepts & Architecture

Guides

Reference

​Overview

​Transit Gateway Performance

​High-Performance Transit Gateways

​Spoke Gateway Performance

​Application-Optimized Performance

​Edge Gateway Performance

​VPN and Security Performance

​Performance Optimization Guidelines

​Instance Sizing Recommendations

Transit Gateway Sizing

Spoke Gateway Sizing

Edge Gateway Sizing

Performance Monitoring

​Performance Tuning Tips

Overview

Transit Gateway Performance

High-Performance Transit Gateways

Spoke Gateway Performance

Application-Optimized Performance

Edge Gateway Performance

VPN and Security Performance

Performance Optimization Guidelines

Instance Sizing Recommendations

Performance Tuning Tips