DDoS Protection and Acceleration for Card and Competitive Games

Four routes account for most leaks: fallback addresses hard-coded in the client, latency-test or announcement endpoints hitting the origin directly, mail and third-party callbacks exposing the egress IP, and historical DNS records. Onboarding addresses all four: every public address collapses onto the protected entry, origin egress moves to a separate NAT address, and rotating the IP invalidates the DNS history. After that the origin is effectively unreachable from the public internet — run any new endpoint through the same checklist.

It depends on whether that protection shape is enough. A protected server concentrates scrubbing in one facility, so capacity is capped there and a single public IP stays exposed — sustained pressure on that address often ends in a facility-level blackhole. Game Shield is distributed: traffic spreads across multiple protected nodes, the real origin IP is hidden, and an affected node can be swapped out in seconds. The two are not mutually exclusive; a common setup keeps the origin on the protected server as a last line of defence while Game Shield fronts player connections, giving you both facility-grade mitigation and distributed absorption with nearby routing.

Yes. Card and competitive titles are protected at layer 4: TCP/UDP traffic is forwarded transparently by port without inspecting your payload, so proprietary binary protocols, encrypted custom packets, WebSocket sessions and voice UDP streams all work unchanged — not one line of server code. All you supply is the port range to forward. If you also expose HTTP endpoints for login or top-ups, a layer 7 WAF and rate-limiting rules can be layered on top of those specifically.

Yes. Mitigation capacity is distributed, so attack traffic is absorbed across many sites instead of converging on one entry point — no capacity request needed in advance. Send us the schedule three to five days ahead and we will reserve scrubbing headroom, spread players across multiple entry points, and tune protection tiers and rate limits to your projected concurrency. We can staff the event window for minute-level response and provide an attack and mitigation report afterwards.

Normally no — most customers see latency fall. Players previously took a cross-network detour to a single origin; afterwards they connect to a nearby edge node and traffic reaches the origin over our private backbone and CN2 routes, cutting hops and congestion points. What does add latency is an entry point too far from your players, which is why we benchmark from your actual player regions before choosing the ingress mix instead of handing over one default address.

Multi-carrier ingress plus routing policy. Each region offers entry points on every major carrier and the client picks the best one from live probing; overseas players enter at a nearby node and reach the origin over the backbone. Players in the same room can also be pinned to a shared entry so their return paths match and the latency gap at one table narrows. The console breaks down latency, jitter and loss by region and carrier, so a complaint can be traced to the access leg or the origin leg quickly.