MaaS Inference Sizer
Size GPU capacity, VRAM, KV cache, TTFT, RPS, concurrent requests and high availability (HA) requirements.
Access all technical sizing tools for AI Factory, MaaS, cloud infrastructure and data center architecture from one unified workspace.
Technical tools for architecture planning, capacity sizing and service cost modeling.
Size GPU capacity, VRAM, KV cache, TTFT, RPS, concurrent requests and high availability (HA) requirements.
Plan model training, GPU pools, optimizer memory, parallelism, serving capacity and high availability.
Design storage capacity, throughput, IOPS, access protocols and storage tiers for training and MaaS workloads.
Plan racks, power distribution, cooling, facility capacity, energy consumption and physical data center infrastructure.
Size CPU, memory, SAPS, SAN, NFS, local disks and mount layouts for SAP HANA Scale-Up and Scale-Out.
Calculate CAPEX, OPEX, licensing and unit service costs for IaaS, PaaS and GPUaaS offerings.
Plan WAN latency, TCP/UDP throughput, BDP, file transfer time, network capacity and SLA targets.
Plan RoCEv2 fabrics, GPU rails, leaf/spine switches, ports, optics, fiber, OOB and VLAN/CIDR capacity for AI factories. 100G–1.6T networking.
Size optical modules, OM3/OM4/OM5/OS2 fiber, cables, strands, patch panels and loss budgets for 1G–1.6T Ethernet. Plan multiple connection groups.