iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
128K scale is an entry-count specification, not a speed rating. It usually means that a named hardware forwarding resource is designed to hold about 128,000 records. The record might be an IPv4 host route, an IPv6 route, a MAC address, an ARP or Neighbor Discovery entry, or an ACL rule. You must identify the table name, switch model, software release, and operating mode before the number has a useful meaning.
For example, Cisco documents 128,000 IPv4 host routes for a Nexus 9364C when its system-routing Layer 3 scale mode is enabled. The same switch is listed at 96,000 IPv4 host routes in the default system-routing mode. Those figures describe forwarding-table capacity; they do not describe port bandwidth, packet rate, latency, or buffer size.
What the “K” counts
In networking specifications, “K” is shorthand for thousand. A 128K scale statement therefore describes approximately 128,000 entries in a particular table. It is a capacity for forwarding state stored in hardware, normally in an ASIC or a related forwarding resource. The switch can consult those entries at line rate when packets arrive.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
The number is meaningful only with its label. “128K IPv4 host routes” is a statement about one type of route entry. “128K MAC addresses” is a statement about Layer 2 address-learning capacity. They are not interchangeable pools simply because both use the same 128K notation. Vendors may also publish separate figures for IPv4 longest-prefix-match (LPM) routes, IPv6 routes, ARP or ND neighbors, multicast groups, and access-control entries.
#1 Best Overall
- 𝗢𝗻𝗲 𝗦𝘄𝗶𝘁𝗰𝗵 𝗠𝗮𝗱𝗲 𝘁𝗼 𝗘𝘅𝗽𝗮𝗻𝗱 𝗡𝗲𝘁𝘄𝗼𝗿𝗸: 5× 10/100/1000Mbps RJ45 Ports supporting Auto Negotiation and Auto MDI/MDIX.
- 𝗚𝗶𝗴𝗮𝗯𝗶𝘁 𝘁𝗵𝗮𝘁 𝗦𝗮𝘃𝗲𝘀 𝗘𝗻𝗲𝗿𝗴𝘆: Latest innovative energy-efficient technology greatly expands your network capacity with much less power consumption and helps save money.
- 𝗥𝗲𝗹𝗶𝗮𝗯𝗹𝗲 𝗮𝗻𝗱 𝗤𝘂𝗶𝗲𝘁: IEEE 802.3X flow control provides reliable data transfer and Fanless design ensures quiet operation.
- 𝗣𝗹𝘂𝗴 𝗮𝗻𝗱 𝗣𝗹𝗮𝘆: Easy setup with no software installation or configuration needed.
- 𝗔𝗱𝘃𝗮𝗻𝗰𝗲𝗱 𝗦𝗼𝗳𝘁𝘄𝗮𝗿𝗲 𝗙𝗲𝗮𝘁𝘂𝗿𝗲𝘀: Prioritize your traffic and guarantee high quality of video or voice data transmission with Port-based 802.1p/DSCP QoS and IGMP Snooping.
128K scale is not 128 Kb/s
A lowercase b in Kb/s denotes bits per second, a bandwidth unit. A capital K in a scale figure normally denotes a count of entries. A switch with a 400GbE interface can move traffic at very high throughput while having a route table smaller than, equal to, or larger than 128,000 entries. Conversely, a device with a large route table may have comparatively modest interface speeds.
Evaluate these characteristics independently:
- Port bandwidth: the nominal speed of each interface and the aggregate capacity of the switch.
- Packet-processing rate: packets per second, especially important for small packets.
- Latency: forwarding delay under the stated test or operating conditions.
- Buffering: memory available for congestion and burst absorption.
- Forwarding-table capacity: the number and type of entries that hardware can install.
A datasheet that highlights 128K does not establish any of the other four performance characteristics.
Which table does 128K refer to?
Use the exact resource name from the specification. Common labels include:
Rank #2
- GIGABIT ETHERNET PORTS: Features 5 x 1.0Gbps Ethernet ports for high-speed connectivity. Auto-negotiating ports detect the optimal speed for connected devices and work with existing Cat5e or Cat6 Ethernet cables.
- PLUG-AND-PLAY UNMANAGED NETWORK SWITCH: Simple plug-and-play setup with no software to install or configuration required.
- FLEXIBLE MOUNTING OPTIONS: Compact metal design supports desktop or wall-mount placement for versatile installation.
- SILENT & ENERGY-EFFICIENT OPERATION: Fanless design ensures silent performance, while IEEE 802.3az Energy Efficient Ethernet reduces power consumption without compromising high-speed network performance.
- REGIONAL COMPATIBILITY: Made for use in U.S. & CA only
| Resource label | What one entry represents | Why it matters |
|---|---|---|
| IPv4 host routes | A route to an individual IPv4 address, typically a /32 | Determines how many individual IPv4 destinations can be installed in the relevant host-route table. |
| IPv4 LPM routes | A prefix used for longest-prefix-match forwarding | Represents network prefixes such as /24 or /16 rather than only individual hosts; the capacity can be separate from host-route capacity. |
| IPv6 routes | An IPv6 prefix or host route | IPv6 entries may use a different hardware table or consume resources differently from IPv4. |
| MAC addresses | A learned or configured Layer 2 address | Limits the number of Ethernet endpoints the switching table can represent. |
| ARP or ND entries | An IPv4 ARP binding or IPv6 Neighbor Discovery binding | Maps a network-layer address to a Layer 2 next hop; this is distinct from the route itself. |
| ACL or policy entries | A match-and-action rule | Security and policy features consume separate or shared ternary-content-addressable memory and can affect other capacities. |
Do not add capacities from different rows as if they were one universal 128,000-entry bucket. The implementation may provide independent tables, or several features may draw from a shared hardware pool.
The Cisco Nexus 9364C example
Cisco’s Cisco Nexus 9000 NX-OS Verified Scalability Guide, Release 10.4(1)F lists two IPv4 host-route figures for the Nexus 9364C. The difference is the configured system-routing mode:
| Nexus 9364C routing mode | Documented IPv4 host-route scale |
|---|---|
| Default system routing mode | 96,000 IPv4 host routes |
| System routing Layer 3 scale enabled | 128,000 IPv4 host routes |
These values are platform- and release-specific. They are not an industry-wide promise that every data-center switch supports 128,000 routes. The 128,000 figure applies to the named model with the stated Layer 3 scale configuration in that guide. A different Nexus model, a different hardware revision, or another NX-OS release can publish different limits.
Rank #3
- GIGABIT ETHERNET PORTS: Features 8 x 1.0Gbps Ethernet ports for high-speed connectivity. Auto-negotiating ports detect the optimal speed for connected devices and work with existing Cat5e or Cat6 Ethernet cables.
- PLUG-AND-PLAY UNMANAGED NETWORK SWITCH: Simple plug-and-play setup with no software to install or configuration required.
- FLEXIBLE MOUNTING OPTIONS: Compact metal design supports desktop or wall-mount placement for versatile installation.
- SILENT & ENERGY-EFFICIENT OPERATION: Fanless design ensures silent performance, while IEEE 802.3az Energy Efficient Ethernet reduces power consumption without compromising high-speed network performance.
- REGIONAL COMPATIBILITY: Made for use in U.S. & CA only
Why a configuration mode changes the number
Switch ASICs have finite on-chip memory and lookup structures. A vendor can expose different “templates” or system modes that allocate those structures for different workloads. A default template may reserve more space for one class of entries or features, while a Layer 3 scale template assigns more resources to routing.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallChanging the mode is therefore a resource-allocation decision, not a software switch that creates new physical memory. It can alter the maximum number of route, host, MAC, ACL, or other entries available to the system. The exact trade-offs are model- and release-dependent, so read the configuration guide and verified-scalability tables for the SKU you are deploying.
Scope: where is the limit applied?
A scale number can apply at several different levels. Determine the scope before comparing it with your network’s inventory:
Rank #4
- One Switch Made to Expand Network-16× 10/100/1000Mbps RJ45 Ports supporting Auto Negotiation and Auto MDI/MDIX
- Gigabit that Saves Energy-Latest innovative energy-efficient technology greatly expands your network capacity with much less power consumption and helps save money
- Reliable and Quiet-IEEE 802.3X flow control provides reliable data transfer and Fanless design ensures quiet operation
- Plug and Play-Easy setup with no software installation or configuration needed
- Advanced Software Features-Prioritize your traffic and guarantee high quality of video or voice data transmission with Port-based 802.1p/DSCP QoS and IGMP Snooping
- ASIC: each forwarding chip may have its own table and limit. A chassis total can look larger while one ASIC is already full.
- Line card: modular systems may publish capacity per line card rather than for the entire chassis.
- Forwarding instance: a virtual routing and forwarding (VRF) instance can have a limit, while the platform also enforces an aggregate limit across all VRFs.
- Chassis aggregate: the headline number may apply only after resources from multiple forwarding components are combined.
Ask the vendor whether the figure is per ASIC, per port group, per line card, per VRF, or an aggregate. A design that fits the aggregate number can still fail if routes are unevenly distributed across hardware partitions.
Feature interactions can reduce usable scale
Published maxima are usually measured for a defined feature set. Real deployments often enable several features at once, and those features may share forwarding memory. VXLAN and EVPN overlays, multicast, ACLs, telemetry, encryption, QoS classification, and other services can consume entries or lookup resources that are also needed by routing.
For that reason, treat a 128K figure as a documented ceiling under stated conditions, not as a guaranteed number of free route slots after every feature is enabled. Check the platform’s resource-utilization commands and the release-specific scale guide after applying your intended configuration. If the documentation provides separate limits for a combined feature set, use those combined figures instead of adding isolated maximums.
Best Value
- 8 GIGABIT PORTS: Features 8 RJ45 ports supporting 10/100/1000 Mbps speeds, providing high-speed wired network connectivity for computers, printers, gaming consoles, and other Ethernet-enabled devices
- PLUG AND PLAY SETUP: No configuration required; simply connect the switch to your network devices and it is ready to use immediately, making network expansion quick and hassle-free
- FANLESS QUIET DESIGN: The fanless design ensures silent operation, making this switch suitable for noise-sensitive environments such as home offices, bedrooms, or conference rooms
- STURDY METAL CONSTRUCTION: Built with a durable metal housing and shielded ports that provide reliable performance, better heat dissipation, and protection against electromagnetic interference
- TRAFFIC OPTIMIZATION: Supports IEEE 802.3x flow control and advanced traffic optimization technology to reduce data bottlenecks and ensure smooth, efficient data transfer across your network
How to read a 128K claim in a datasheet
- Copy the complete label. Record whether the claim says IPv4 host routes, IPv4 LPM, IPv6, MAC, ARP/ND, ACL, or another resource.
- Identify the exact hardware SKU. “Nexus 9000” is a family name; the 9364C and other members can have different tables.
- Check the software release. Verified scale tables are tied to an NX-OS release. Use the release named in the document or verify the current release notes.
- Find the operating mode. Note whether the value requires a default template, a Layer 3 scale template, or another profile.
- Confirm the scope. Look for per-ASIC, per-line-card, per-VRF, and chassis-wide qualifiers.
- List enabled features. Include overlays, multicast, security policy, telemetry, encryption, and QoS before estimating remaining capacity.
- Compare with live counters. On a deployed switch, inspect hardware resource utilization and route-installation status rather than relying only on the datasheet maximum.
- Leave operational headroom. Size for expected growth, failover events, and temporary control-plane churn; do not plan to operate exactly at the documented ceiling.
What happens when the table fills?
Control-plane protocols can learn more routes than the forwarding hardware can install. When that happens, the switch may keep some routes in software or reject additional entries, depending on the platform and feature. Packets whose destinations lack a hardware entry can be dropped, punted to a slower path, or forwarded according to a less-specific route. The visible symptom might be selective reachability loss rather than a total outage.
Investigate in this order:
- Check whether the control plane learned the route and whether the hardware installed it.
- Identify which resource is exhausted; a full ACL or MAC table can coexist with unused IPv4 host-route slots.
- Review the active system template and confirm that it is the intended one.
- Look for an uneven ASIC or VRF partition, where one slice is full before the aggregate limit is reached.
- Examine recent feature changes, such as an EVPN deployment, new telemetry rules, or expanded policy, that may have consumed shared resources.
- Reduce or summarize entries, redistribute state across hardware partitions, or move to a platform with a larger documented table when the design genuinely exceeds the limit.
A practical leaf-switch example
Imagine a leaf receiving host routes from a control-plane protocol. A specification of 128K IPv4 host-route scale means the relevant forwarding hardware is documented for approximately 128,000 individual IPv4 host-route entries under the specified mode. It does not say how many MAC addresses the leaf can learn, how many IPv6 prefixes it can hold, or how fast its uplinks transmit.
Before approving the design, map each requirement to a resource: host routes for endpoint reachability, LPM routes for summarized networks, MAC entries for Layer 2 endpoints, ARP/ND for next-hop resolution, and ACL entries for policy. Then verify that the selected mode and enabled features leave capacity in every resource, not just the one highlighted in the product brief.
Questions to ask before comparing switches
- Is the quoted number for host routes or prefix routes?
- Does it apply to IPv4, IPv6, or both, and are the tables separate?
- Is the number per ASIC, per forwarding instance, or an aggregate?
- Which configuration template and software release are required?
- How do VXLAN/EVPN, multicast, ACLs, telemetry, encryption, and QoS change the available resources?
- What are the independent throughput, packet-rate, latency, and buffer specifications?
- Which operational commands show current hardware utilization and failed route installations?
Answering those questions turns a marketing-scale label into a capacity figure you can actually use for design and operations.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
