Unlike bandwidth, latency and jitter depend on the specific context of network topology and traffic conditions. In particular, for automatic algorithmic trading, latency variability needs to be monitored with the same diligence that traders monitor changing market prices, correlations, and trends. Otherwise, the variability in network performance could drown out the sophistication of the trading algorithms. Spending money on expensive equipment, bandwidth, and buzzwords with no understanding of the underlying trade-offs is not an effective latency engineering strategy and will lead to poor results.

Download the whitepaper Pragmatic Network Latency Engineering (pdf).