AI workloads are filling data centres with traffic, much of it encrypted, and the usual way to keep it moving has been to switch off advanced threat prevention or buy far more firewall hardware than a network needs most of the time. Check Point Software Technologies says its new Maestro Hyperscale Orchestrator 200 (MHO 200) removes that trade-off, and it is available now in all regions, Malaysia included.
Quick facts
- Total fabric capacity of 12.8 Tbps, up from 3.2 Tbps on the MHO 175, in the same 1RU size
- Up to 4.5 Tbps of threat prevention throughput per Maestro system, up from 1.1 Tbps
- New 400GbE ports, the speed high-end data centres and AI facilities are standardising on
- Automatic failover across two sites for 99.999% uptime, with sessions kept live through failures
- On Check Point's product catalogue for all regions; no ringgit pricing published
What Maestro does
Maestro is an orchestrator: it ties several Check Point firewalls, including different models and generations, into one system that grows as demand rises, instead of a fixed chassis tied to dedicated hardware. That system can also be split into up to eight independent security groups, each with its own rules, and teams manage the whole thing as a single object. Check Point's Agentic Network Security Orchestration then automates policy tuning, health monitoring and compliance across it.
The MHO 200 is the part that sits at the front. Its job is to spread traffic across the firewalls behind it, and the jump to 400GbE ports is what lifts the ceiling four times over the MHO 175.
Who it is for
Check Point names large enterprises modernising their data centres, neoclouds (cloud providers built to rent out GPU capacity for AI training and inference), colocation operators, telcos protecting their backbone and governments building sovereign AI infrastructure. In Malaysia, where new data centres keep being announced, that list describes the kind of operators building them.
On the AI side, the orchestrator lets teams find and secure AI applications, large language models, agents and Model Context Protocol (MCP) servers directly from the firewall, building on the AI Network Firewall Check Point released in July. For private and sovereign AI projects, its AI Factory Firewall works with NVIDIA BlueField data processing units, and live patching applies security updates without reboots.
"Organizations should not have to choose between inspecting that traffic and keeping it moving," said Pete Finalle, Research Manager, Security and Trust at IDC, who expects security platforms that scale threat prevention without overprovisioning hardware to be essential for neoclouds, telcos and enterprises building AI infrastructure.
Price and availability
The MHO 200 is listed on Check Point's product catalogue for all regions now. Check Point has not published pricing, which is normal for enterprise network hardware sold through partners and quoted per deployment. Businesses that would rather not buy hardware outright can look at the company's three-year firewall subscription launched on 1 November. Full specifications are on Check Point's Maestro page.
