Switches
The switching backbone for data-center and enterprise networks.
Switching hardware built for line-rate, low-latency fabrics.
Our top-of-rack and aggregation switches deliver the non-blocking throughput and deterministic latency modern data centres demand, with high-density ports and robust thermals. Manufactured in India, they give operators a supported, locally sourced foundation for their network fabric.
From the rack to the core, non-blocking throughout.
25G Access
Top-of-Rack Switch
A leaf switch presenting 48 server-facing 25GbE ports with 100GbE uplinks, giving a clean, non-oversubscribed access layer that scales into 100/400G spine tiers as the fabric grows.
100G Aggregation
Spine Switch
A high-radix 100GbE aggregation switch forms the non-blocking spine of a leaf-spine fabric, delivering full line-rate ECMP forwarding for east-west traffic across the data center.
Campus Network
Enterprise Switch
A managed L2/L3 campus switch with rich access-layer features and PoE options, connecting users, wireless and IoT endpoints on a supported, locally manufactured platform.
Non-Blocking
High-Density Ports
A wire-speed fabric provides up to 3.6 Tbps of aggregate bandwidth with full line-rate forwarding on every port simultaneously, so there is no internal congestion under all-to-all traffic.
Hot-Swap PSU
Redundant Power
Dual load-sharing, hot-swappable power supplies and N+1 fan trays keep the switch forwarding through a component failure or field service, with no maintenance window or fabric outage.
Fabric Integration
Data-Center Ready
Support for open network operating systems, VXLAN/EVPN overlays and automation APIs lets operators manage the fabric as code with zero-touch provisioning and streaming telemetry.
The fabric that never blocks.
25G Access, 100G Uplink
Top-of-rack switches present 48 × 25GbE server-facing ports with 6 × 100GbE uplinks, giving a clean non-oversubscribed leaf that scales into 100/400G spine layers as the fabric grows.
Non-Blocking Throughput
A wire-speed switching fabric delivers up to 3.6 Tbps of aggregate bandwidth with full line-rate forwarding on every port simultaneously, with no internal congestion, even under all-to-all traffic.
Deterministic Low Latency
Cut-through forwarding holds port-to-port latency in the sub-microsecond range, meeting the tight, predictable timing that distributed storage, HPC and AI collective traffic depend on.
Full L2 / L3 Feature Set
Line-rate L2 switching and L3 routing with VXLAN/EVPN overlays, BGP, OSPF and ECMP build modern spine-leaf underlays and multi-tenant overlays on a single, programmable platform.
Redundant Hot-Swap Power
Dual load-sharing, hot-swappable power supplies and N+1 fan trays keep the switch forwarding through a component failure or field service, with no maintenance window and no fabric outage.
Open, Programmable NOS
Support for open network operating systems and automation APIs lets operators manage the fabric as code, with zero-touch provisioning, streaming telemetry and CI-driven configuration at scale.
- Port Speeds
- 1 / 10 / 25 / 100 GbE
- ToR Config
- 48× 25G + 6× 100G
- Switching Capacity
- Up to 3.6 Tbps
- Latency
- Sub-microsecond
- Forwarding
- Non-blocking, cut-through
- Protocols
- L2 / L3, VXLAN/EVPN, BGP
- Power
- Redundant hot-swap PSU
- Cooling
- N+1 hot-swap fans
- Origin
- 100% Made in India