Network Packet Broker 8x 100G QSFP28 with 48x 25G SFP28 Ports
Network Packet Broker 8x 100G QSFP28 with 48x 25G SFP28 Ports
PX308P-48Y-T
-
Support L2 to L7 Traffic Aggregation, Filtering, Replication, Load Balancing, Correlation;
-
Provide up to 8x100G QSFP28 with 44x25G/10G, 4x25G/10G/1G ports;
-
Tunnel preprocessing, full tunnel inner layer IP 5-tuple filtering, load balancing, and tunnel stripping;
-
Support packet deduplication, support IPv4 and IPv6;
-
Time Stamps, Editing VLAN of Packets, Packet Truncation, Header Modification;
-
Support Web, SNMP, SNMP Trap, REST AP;
-
Packet preprocessing includes time stamping, packet slicing, VLAN tagging, MAC modification, and GRE termination;
Network Packet Broker 8x 100G QSFP28 with 48x 25G SFP28 Ports, 3.3Tbps Switching Capacity
The Asterfusion Network Packet Broker is made to make your tracking and security tools more powerful. This packet broker has 48x 25G SFP28(or 48x 10G SFP+) ports for aggregating all kinds of traffic and 8x 100G QSFP28(or 8x 40G QSFP+) ports for sending data quickly. You don’t have to worry about packets getting lost because of the blazing-fast 3.3Tbps switching capacity. Due to it being based on the Barefoot Tofino chip, this Network Packet Broker is fast and reliable. Its PX308P-48Y-T design with 2 DPUs also lets you customization features.
Specifications
|
Quality Certifications




Features
Basic feature
- N-tuple rules with mask (IPv4/IPv6 5-tuple, MAC, Ether Type, VLAN tags, VNI, TCP Flag, DSCP, Fragment message, etc.)
- Exact IPv4/IPv6 5-tuple rules
- Based on MAC information rules, source MAC, destination MAC
- Inner 5-tuple rules with mask
- Inner exact 5-tuple rules
- Supports multi-level concurrency strategies, can execute different traffic strategies on the inbound, intermediate, and outbound interfaces of the pipeline, and supports L2/L3/L4/L7 layer combination policy rules
- Signature rules, payload partial match depth 40 bytes
- Forwarding by flow classification
- Forwarding by the specified output port
- Load-balancing output by specified output link group
- Supports normal forwarding of error packets
- Load balancing for policies such as IP 5-tuple, source/destination IP, source/destination IP, source IP/source port, destination IP/ destination port, source IP/destination IP/ source port/destination port, source/destination MAC, etc.
- Load balancing based on inner IP 5-tuple hashing in case of tunnel encapsulation.
- Fault-tolerant load balancing protection mechanism, supporting static load balancing, dynamic load balancing, and elastic load balancing mechanisms.
- Load balancing strategy with weights.
- Multi-level load with flexible and configurable hash seeds.
- Load balanced output within a group after traffic replication.
- Supports fragmentation following function, ensuring that the first piece and non-first piece of the same IP packet are output from the same port.
-
Load balanced output from the same source and same destination.
- VLAN tag adding, modifying, deleting
- ns timestamp
- Tagging source port
- Packet slicing, you can set the reserved byte length which based on ACL rules
- BPDU packets are forwarded by default, and BPDU packets can also be discarded
- Source/destination MAC address modification
- GRE encapsulation (supporting configuration with IP and MAC address)
- Tunnel encapsulation stripping for GTP, GRE, VXLAN, VXLAN, ERSPAN, CFP, IP over IP, SRv6, MPLS, MPLS in UDP, MPLS in GRE, etc.
- GRE termination (Each port can be configured with IP address)
- Support interface master and link group master and backup: when the main port fails, the traffic will be automatically switched to the backup port for output, meanwhile, it supports an automatic recovery mechanism.
- Supports interface linkage and link group linkage: when an interface fails, all the interfaces linked to it will be synchronized down, and supports an automatic recovery mechanism.
- Supports the protection mechanism of single link or link group as backup link between devices, and supports an automatic recovery mechanism.
- When networking across data centers, it supports device self-test, device mutual test, link detection between devices, and cross-device link detection.
Port feature
Management feature
Data Processing Unit Module (Optional)


DPU model: Marvell Octeon, 24-core ARM64, 1.8GHz
Memory: 32GB DDR4, maximum 96GB
Flash: 64GB eMMC 5.1
SSD (optional): Expandable to 2xM.2 NVME
Advanced feature (DPU support)
- Signature rules (supports signature based on L2/L3/L4 fixed offset, signature with fixed position, signature with floating window range)
- Specific URL (matching based on the complete host field and the specific URL)
- Combination rules
- Net flow traffic statistics
Flow sampling (sampling can be performed according to the 5-tuple of the hit rule, etc., and the sampling rate can be set) - Header output, only forwards the header of the network layer and transport layer of the original packets
- Message deduplication (support sliding window settings, support flexible settings to ignore comparison item deduplication)
- IP fragment reorganization (supports IP slice following, synthesized packet output and original message discard function)
- TCP out-of-order reorganization (supports bi-directional out-of-order reorganization for each session)
- TCP encapsulation (supports encapsulation using TCP, supports encapsulation using upper layer SDTP protocol)
- Message desensitization (supports keywords or intervals of the set number and encryption desensitization)
- Support SSL3.0, TLS1.0, TLS1.1, and TLS1.2 protocol decryption (optional features) Support RSA, ECDH asymmetric encryption and decryption algorithms; Support AES_CBC, AES_GCM, 3DES symmetric encryption, and decryption algorithms; support SHA1, SHA256, SHA384 hash algorithms
- Load balancing based on polling, weighting mode
Connectivity Solutions
