HPE JH321A Getting Started Guide
HPE FlexFabric 5950
Switch Series
Data sheet
Key features
• Cut-through with ultra-low-latency and
wire speed
• VXLAN, VTEP, and OVSDB support for
virtualized environments
• High-density 100GbE/40GbE/25GbE/
10GbE spine/top-of-rack (ToR) connectivity
• IPv6 support with full L2 and L3 features
• New HPE FlexFabric Network Analytics
solution capability for real time microburst
detection
Product overview
The HPE FlexFabric 5950 Switch Series is a
family of high-density, ultra-low-latency, and
ToR switches that is part of HPE FlexFabric
solution (from the HPE FlexNetwork
architecture).
Ideally suited for deployment at the
aggregation or server access layer of large
enterprise data centers, the HPE 5950
Switch Series is also powerful enough
for deployment at the core layer of
medium-sized enterprises.
With the increase in virtualized applications
and server-to-server traic, customers require
a spine and ToR switches that can meet
their throughput requirements. With the
HPE FlexFabric 5950, Data Center can
now support up to 100G per ports, allowing
high performance server connectivity and the
capabilities to handle virtual environments.
This is available in the low-latency,
all-in-a-single device oering—HPE FlexFabric
5950 Switch Series.
Features and benefits
Quality of service (QoS)
• Powerful QoS features
– Flexible queue scheduling
Including strict priority (SP), WRR, WDRR,
WFQ, SP+WRR, SP+WDRR, SP+WFQ,
configurable buffer, time range, queue
shaping, and CAR with 8 kbps granularity
– Packet filtering and remarking
Packet filtering at L2 through L4; flow
classification based on source MAC
address, destination MAC address, source
IP (IPv4/IPv6) address, and destination IP
Data sheet
Page 2
Data center optimized
• Flexible high-port density
HPE FlexFabric 5950 Series enables
scaling of the server Edge, with 100GbE,
40GbE, 25GbE, and 10GbE Spine and
Leaf deployment. The 5950 series comes
as a high-density 32-port 100G fixed
port, as a 48-port of 25G with 8-port
100G and in a 2RU modular form factor
• High-performance switching
Cut-through and non-blocking architecture
delivers low latency (~1 microsecond for
100GbE) for very demanding enterprise
applications; the switch delivers
high-performance switching capacity and
wire-speed packet forwarding
• Higher scalability
Intelligent Resilient Fabric (IRF) technology
simplifies the architecture of server access
networks; up to 10 HPE 5950 switches
can be combined to deliver unmatched
scalability of virtualized access layer
switches and flatter two-tier networks using
IRF, which reduces cost and complexity
• Advanced modular operating system
Comware v7 software’s modular design
and multiple processes bring native high
stability, independent process monitoring,
and restart; the OS also allows individual
software modules to be upgraded for
higher availability and supports enhanced
serviceability functions such as hitless
software upgrades with IRF-based in
services software upgrade (ISSU)
• Reversible airflow
Enhanced for data center hot-cold aisle
deployment with reversible airflow—for
either front-to-back or back-to-front airflow
• Redundant fans and power supplies
Internal redundant and hot-pluggable
power supplies and dual fan trays enhance
reliability and availability
• Lower OPEX and greener data center
Provides reversible airflow and advanced
chassis power management
• Data center bridging (DCB) protocols
Provides support for IEEE 8021Qbb priority
flow control (PFC), data center bridging
exchange (DCBX), IEEE 8021Qaz enhanced
transmission selection (ETS), explicit
congestion notification (ECN) for converged
FCoE, iSCSI, and RoCE environments
• Jumbo frames
With frame sizes of up to 9416 bytes on
100GbE ports, high-performance remote
backup and disaster-recovery services to
are enabled
• VXLAN hardware support
VXLAN L2 gateway support for up to 2K
tunnels
• Dynamic VXLAN configuration
OVSDB support for dynamic VXLAN
configuration
Manageability
• The HPE FlexFabric Network Analytics
solution with real-time telemetry analysis
provides insight into data center network
operation
– Tracks all the accounting associated with
the admission and allocation process
of all the buers and queues across the
ingress and egress ports
– Microburst congestion detection
– Rich congestion analytics
– Buer congestion state and statistics
– For more information, see the
HPE FlexFabric Network Analytics
data sheet and HPE FlexFabric
Network Analytics white paper.
• Full-featured console
Provides complete control of the switch
with a familiar CLI
• Troubleshooting
– Ingress and egress port monitoring
Enables network problem solving
– Traceroute and ping
Enables testing of network connectivity
• Multiple configuration files
Allows multiple configuration files to be
stored to a flash image
• sFlow® (RFC 3176)
Provides wire-speed traic accounting and
monitoring
• SNMPv1, v2c, and v3
Facilitates centralized discovery, monitoring,
and secure management of networking
devices
• Out-of-band interface
Isolates management traic from user data
plane traic for complete isolation and
total reachability, no matter what happens
in the data plane
• Remote configuration and management
Delivered through a secure CLI over
Telnet and SSH; role-based access
control (RBAC) provides multiple levels of
access; configuration rollback and multiple
configurations on the flash provide ease
of operation; remote visibility is provided
with sFlow and SNMPv1/v2/v3, and
is fully supported in HPE Intelligent
Management Center (IMC)
• ISSU and hot patching
Provides hitless software upgrades with
IRF-based ISSU and hitless patching of the
modular operating system
• NTP support
Synchronizes timekeeping among
distributed time servers and clients;
support for network time protocol (NTP)
Resiliency and high availability
• IRF technology
Enables an HPE FlexFabric to deliver
resilient, scalable, and secured data center
networks for physical and virtualized
environments; groups up to 10 HPE 5950
switches in an IRF configuration, allowing
them to be configured and managed as a
single switch with a single IP address;
simplifies ToR deployment and management,
reducing data center deployment and
operating expenses
• IEEE 8021w Rapid Convergence
Spanning Tree Protocol (STP)
Increases network uptime through faster
recovery from failed links
• IEEE 8021s Multiple Spanning Tree
Provides high-link availability in multiple
VLAN environments by allowing Multiple
Spanning Trees
• Virtual router redundancy protocol (VRRP)
Allows groups of two routers to back each
other up dynamically to create highly available
routed environments
• Hitless patch upgrades
Allows patches and new service features
to be installed without restarting the
equipment, increasing network uptime, and
facilitating maintenance
Data sheet
Page 3
• Fast protocol convergence with
standard-based failure detection—
bidirectional forwarding detection (BFD)
Enables link connectivity monitoring and
reduces network convergence time for RIP,
OSPF, BGP, IS-IS, VRRP, MPLS, and IRF
• Device link detection protocol (DLDP)
Monitors link connectivity and shuts down
ports at both ends if unidirectional traic
is detected, preventing loops in STP-based
networks
• Graceful restart
Allows routers to indicate to others their
capability to maintain a routing table during
a temporary shutdown and significantly
reduces convergence times upon recovery;
supports OSPF, BGP, and IS-IS
L2 switching
• Address resolution protocol (ARP)
Supports static, dynamic, and reverse ARP
and ARP proxy
• IEEE 8023x Flow Control
Provides intelligent congestion
management via PAUSE frames
• Ethernet Link Aggregation
Provides IEEE 8023ad Link Aggregation
of up to 256 groups of 32 ports; support
for LACP, LACP local forwarding first, and
LACP short-time provide a fast, resilient
environment that is ideal for the data center
• Spanning Tree Protocol
Supports STP (IEEE 8021D), Rapid STP
(RSTP, IEEE 8021w), and Multiple STP
(MSTP, IEEE 8021s)
• VLAN support
Provides support for 4096 VLANs based
on port
• IGMP support
Provides support for IGMP Snooping,
fast-leave, and group policy; IPv6 IGMP
Snooping provides L2 optimization of
multicast traic
• DHCP support at L2
Provides full DHCP Snooping support for
DHCP Snooping Option 82, DHCP Relay
Option 82, DHCP Snooping Trust, and
DHCP Snooping Item Backup
L3 services
• Address resolution protocol
Determines the MAC address of another
IP host in the same subnet; supports static
ARPs; gratuitous ARP allows detection of
duplicate IP addresses; proxy ARP allows
normal ARP operation between subnets
or when subnets are separated by a
L2 network
• Dynamic host configuration protocol
Simplifies the management of large IP
networks and supports client and server;
DHCP Relay enables DHCP operation
across subnets
• Operations, administration, and
maintenance (OAM) support
Provides support for Connectivity Fault
Management (IEEE 8021AG) and Ethernet
in the First Mile (IEEE 8023AH); provides
additional monitoring that can be used for
fast fault detection and recovery
L3 routing
• VRRP and VRRP extended
Allows quick failover of router ports
• Policy-based routing
Makes routing decisions based on policies
set by the network administrator
• Equal-cost multipath (ECMP)
Enables multiple equal-cost links in a
routing environment to increase link
redundancy and scale bandwidth
• L3 IPv4 routing
Provides routing of IPv4 at media speed;
supports static routes, RIP and RIPv2, OSPF,
BGP, and IS-IS
• Open shortest path first
Delivers faster convergence; uses this
link-state routing interior gateway protocol
(IGP), which supports ECMP, NSSA, and MD5
authentication for increased security and
graceful restart for faster failure recovery
• Border Gateway Protocol 4 (BGP-4)
Delivers an implementation of the EGP
utilizing path vectors; uses TCP for enhanced
reliability for the route discovery process;
reduces bandwidth consumption by
advertising only incremental updates;
supports extensive policies for increased
flexibility; scales to very large networks
• Intermediate system to intermediate
system (IS-IS)
Uses a path vector IGP, which is defined by
the ISO organization for IS-IS routing and
extended by IETF RFC 1195 to operate in
both TCP/IP and the OSI reference model
(Integrated IS-IS)
• Static IPv6 routing
Provides simple manually configured IPv6
routing
• Dual IP stack
Maintains separate stacks for IPv4
and IPv6 to ease the transition from
an IPv4-only network to an IPv6-only
network design
• Routing Information Protocol next
generation (RIPng)
Extends RIPv2 to support IPv6 addressing
• OSPFv3
Provides OSPF support for IPv6
• BGP+
Extends BGP-4 to support multiprotocol
BGP (MBGP), including support for IPv6
addressing
• IS-IS for IPv6
Extends IS-IS to support IPv6 addressing
• IPv6 tunneling
Allows IPv6 packets to traverse IPv4-only
networks by encapsulating the IPv6 packet
into a standard IPv4 packet; supports
manually configured, 6 to 4, and intra-site
automatic tunnel addressing protocol
(ISATAP) tunnels; is an important element for
the transition from IPv4 to IPv6
• Policy routing
Allows custom filters for increased
performance and security; supports ACLs,
IP prefix, AS paths, community lists, and
aggregate policies
• Bidirectional forwarding detection (BFD)
Enables link connectivity monitoring and
reduces network convergence time for RIP,
OSPF, BGP, IS-IS, VRRP, MPLS, and IRF
• Multicast Routing PIM dense and
sparse modes
Provides robust support of multicast
protocols
Data sheet
Page 4
• L3 IPv6 routing
Provides routing of IPv6 at media speed;
supports static routing, RIPng, OSPFv3,
BGP4+ for IPv6, and IS-ISv6
Additional information
• Green IT and power
Improves energy eiciency through
the use of the latest advances in silicon
development; shuts o unused ports and
utilizes variable-speed fans, reducing
energy costs
Management
USB support
• File copy
Allows users to copy switch files to and
from a USB flash drive
• Multiple configuration files
Stores easily to the flash image
• SNMPv1, v2c, and v3
Facilitates centralized discovery, monitoring,
and secure management of networking
devices
• Out-of-band interface
Isolates management traic from user data
plane traic for complete isolation and
total reachability, no matter what happens
in the data plane
• Port mirroring
Enables traic on a port to be simultaneously
sent to a network analyzer for monitoring
• Remote configuration and management
Is available through a CLI
• IEEE 8021AB Link Layer Discovery
Protocol (LLDP)
Advertises and receives management
information from adjacent devices on
a network, facilitating easy mapping by
network management applications
• sFlow (R FC 3176)
Provides scalable ASIC-based wire-speed
network monitoring and accounting with
no impact on network performance; this
allows network operators to gather a
variety of sophisticated network statistics
and information for capacity planning and
real-time network monitoring purposes
• Command authorization
Leverages RADIUS to link a custom
list of CLI commands to an individual
network administrator’s login; an audit trail
documents activity
• Dual flash images
Provides independent primary and
secondary operating system files for
backup whil e upgrading
• Command-line interface
Provides a secure, easy-to-use CLI for
configuring the module via SSH or a switch
console; provides direct real-time session
visibility
• Logging
Provides local and remote logging of
events via SNMP (v2c and v3) and syslog;
provides log throttling and log filtering
to reduce the number of log events
generated
• Management interface control
Provides management access through
a modem port and terminal interface, as
well as in-band and out-of-band Ethernet
ports; provides access through terminal
interface, Telnet, or SSH
• Industry-standard CLI with a
hierarchical structure
Reduces training time and expenses, and
increases productivity in multivendor
installations
• Management security
Restricts access to critical configuration
commands; oers multiple privilege levels
with password protection; ACLs provide
Telnet and SNMP access; local and remote
syslog capabilities allow logging of all access
• Information center
Provides a central repository for system
and network information; aggregates all
logs, traps, and debugging information
generated by the system and maintains
them in order of severity; outputs the
network information to multiple channels
based on user-defined rules
• Network management
HPE IMC centrally configures, updates,
monitors, and troubleshoots
• Remote intelligent mirroring
Mirrors ingress/egress ACL-selected traic
from a switch port or VLAN to a local
or remote switch port anywhere on the
network
Security
• Access control lists
Provides IP L3 filtering based on
source/destination IP, address/subnet,
and source/destination TCP/UDP port
number
• RADIUS/TACACS+
Eases switch management security
administration by using a password
authentication server
• Secure Shell
Encrypts all transmitted data for secure
remote CLI access over IP networks
• IEEE 8021X and RADIUS network
logins
Controls port-based access for
authentication and accountability
• Port security
Allows access only to specified MAC
addresses, which can be learned or
specified by the administrator
Convergence
• LLDP-MED (Media Endpoint Discovery)
Defines a standard extension of LLDP that
stores values for parameters such as QoS
and VLAN to configure network devices
such as IP phones automatically
Warranty and support
• 1-year warranty
See hpe.com/networking/
warrantysummary for warranty and
support information included with your
product purchase
• Software releases
To find software for your product, refer to
hpe.com/networking/support; for details
on the software releases available with
your product purchase, refer to hpe.com/
networking/warrantysummary
Loading...