Performance
Link Aggregation Calculator
Calculate aggregate bandwidth, redundancy benefits, and load balancing for bonded links. Plan LACP configurations and understand hash algorithm impacts.
Link Configuration
Load Balancing Algorithm
XOR of source and destination IP
Number of unique traffic flows (affects load distribution)
Link Bundle Visualization
Switch A
Switch B
Active (4)
Aggregate Bandwidth
Active Bandwidth
0 Mbps
4x 10.0 Gbps links
Total Capacity
0 Mbps
4 links
Redundant Links
0
standby
Single Flow Max
0 Mbps
per flow limit
Failover Capacity
0 Mbps
after 1 failure
No Redundancy
All links are active. A single link failure will reduce capacity by 10.0 Gbps.
LACP Modes
Active Mode
Initiates LACP negotiation. Sends LACPDU every 1s (fast) or 30s (slow).
Passive Mode
Responds to LACP but doesn't initiate. Requires active peer.
Static (On)
No LACP negotiation. Both sides must be configured identically.
Best Practices
- Use LACP (802.3ad) for dynamic negotiation
- Match speed/duplex on all member links
- Use Layer 3+4 hashing for best distribution
- Consider MLAG/vPC for multi-chassis redundancy
Common LAG Configurations
| Config | Links | Bandwidth | Use Case |
|---|---|---|---|
| 2x 1G | 2 | 2 Gbps | Server uplinks, SMB |
| 4x 1G | 4 | 4 Gbps | Core switch uplinks |
| 2x 10G | 2 | 20 Gbps | Data center servers |
| 4x 25G | 4 | 100 Gbps | Spine-leaf fabric |
| 2x 100G | 2 | 200 Gbps | Core/backbone |
Hash Algorithm Impact
- Layer 2: Only source/dest MAC. Poor with few MAC addresses.
- Layer 3: Source/dest IP. Better distribution for routed traffic.
- Layer 3+4: IP + ports (5-tuple). Best for multi-flow servers.
- Single flow: Always limited to one link's bandwidth.
Multi-Chassis Options
- Cisco vPC: Virtual Port Channel across Nexus switches
- Arista MLAG: Multi-Chassis Link Aggregation
- Juniper MC-LAG: Multi-Chassis LAG
- HP IRF: Intelligent Resilient Framework