Infrastructure

System Status

Real-time overview of our global routing network and data center operations.

All Systems Operational

Equinix NY4 (New York)

100% Uptime

                                Evaluation API
                                Operational
                            
                                Market Data Feed
                                Operational
                            
                                Internal Latency
                                0.04 ms

Equinix LD4 (London)

100% Uptime

                                Evaluation API
                                Operational
                            
                                Market Data Feed
                                Operational
                            
                                Internal Latency
                                0.05 ms

Equinix TY3 (Tokyo)

99.99% Uptime

                                Evaluation API
                                Operational
                            
                                Market Data Feed
                                Operational
                            
                                Internal Latency
                                0.06 ms

Global Anycast Network

TwoWayMind utilizes BGP Anycast routing to automatically direct your API calls and data requests to the nearest physical data center. This ensures that a trader in London connects directly to LD4, while a trader in New York connects to NY4, completely eliminating cross-atlantic latency hops.

Our core matching engine and risk evaluation algorithms run on bare-metal FPGA hardware, ensuring deterministic execution times regardless of market volatility.

Failover & Redundancy

Every single server rack is configured with N+1 redundancy. If a primary evaluation node experiences a hardware failure or kernel panic, our automated load balancers seamlessly route traffic to the hot-standby node within 0.2 milliseconds.

Market data feeds are ingested via dual-fiber cross-connects (A/B lines) directly from the exchanges. If one line drops packets, the secondary line reconstructs the order book instantly without missing a single tick.

Past Incidents

No incidents reported in the last 90 days. Our distributed infrastructure ensures zero-downtime failovers during maintenance windows.

Service Level Agreements (SLA) & Maintenance

Uptime Guarantees

We stand by the stability of our matching engine and data feeds. Our standard Enterprise SLA guarantees a 99.99% uptime for data pipelines during active market hours. Any downtime exceeding 52.56 minutes per year qualifies clients for tier-based service credits.

In the event of catastrophic regional ISP failures, our BGP routers automatically propagate new routes globally in under 3 seconds, ensuring your trading loops re-establish connections instantly.

Scheduled Maintenance Windows

To deploy core kernel upgrades and FPGA bitstream updates, we conduct routine maintenance on Saturdays between 22:00 and 23:00 UTC.

API Gateway: Remains online; rejects orders with specific maintenance codes.
WebSockets: Gracefully disconnected 5 minutes prior to the window.
Client Portal: Remains fully accessible in read-only mode.