Direct answer: what is the Cisco Meraki MS130-48P?

Why the MS130-48P deserves a model-specific evaluation

The MS130-48P sits in a useful but very specific part of the Cisco Meraki access-switch portfolio. It combines forty-eight conventional 1 GbE copper access ports with PoE and four 1 GbE SFP uplinks. That configuration makes sense for many offices, schools, healthcare sites, retail locations, warehouses and distributed enterprise branches where user devices remain primarily gigabit-capable and where the upstream links do not need multi-gigabit or 10-gigabit bandwidth. It is not simply a bigger version of a small office switch; it is a rack-mounted access-layer platform intended to be part of a managed network architecture.

The 740 W PoE switch budget is one of the strongest reasons to shortlist the MS130-48P. A forty-eight-port access layer often supports a mixed population of devices rather than forty-eight identical clients. A typical cabinet can include desk phones, Wi-Fi access points, security cameras, door controllers, conferencing endpoints and ordinary PCs. Some of these devices consume no PoE at all, while others may draw materially more power during full operation than their idle consumption suggests. A large total budget gives the design team more room to support dense powered-device deployments, but the 30 W per-port limit still matters. High-power endpoints must be checked individually rather than assuming that a 740 W chassis budget guarantees compatibility with every PoE device.

The four 1 GbE SFP uplinks also need careful interpretation. They are valuable when the access switch needs fibre connectivity to an aggregation switch, another communications room or a network core, and Cisco Meraki lists compatible one-gigabit SFP transceiver options for the MS130-48P. However, the uplinks are not SFP+. If an organisation expects a large volume of east-west traffic, high-density wireless traffic, frequent local backups, video distribution or aggressive future bandwidth growth, the upstream bottleneck can become more important than the number of access ports. In that case, the MS130-48X or another model with faster uplinks may deserve comparison.

The Meraki cloud operating model is equally important. This switch is designed to be claimed into a Meraki organisation, added to a Dashboard network, connected to the physical network and then managed through the Dashboard. That model can simplify operations for organisations with multiple sites because configuration, monitoring, firmware management and remote troubleshooting are centralised. It also means the licensing model and Dashboard organisation design are not optional procurement afterthoughts. Hardware, licence and operational model should be selected together.

Verified hardware profile at a glance

Specification values above reflect Cisco Meraki documentation for the MS130-48P. Quotation items such as licence, power cord, SFP modules, patching and installation should be identified separately.

Port architecture: where 48 x 1 GbE is the right answer

Forty-eight Gigabit Ethernet access ports are still a highly practical density for many business networks. Workstations, printers, ordinary IP phones, building controllers and many surveillance cameras operate comfortably within a one-gigabit access connection. The MS130-48P therefore allows an organisation to consolidate a large number of edge devices into a single rack unit without forcing the cost or complexity of multigigabit access ports everywhere. That can be particularly sensible when the existing structured cabling, endpoint capabilities and application profile do not justify faster edge links.

Port count should be planned around active connections plus reserve, not around the number of currently patched cables. A branch with thirty-eight active Ethernet devices may appear to fit easily on a forty-eight-port switch, but planned access points, additional cameras, conference-room systems or growth can consume the remaining ports quickly. Many buyers benefit from keeping a sensible spare-port allowance so expansion does not immediately require another chassis. The exact reserve depends on how frequently the environment changes and whether a second access switch already exists in the cabinet.

The port architecture is also a reminder that endpoint bandwidth and uplink bandwidth are different design problems. Forty-eight one-gigabit edge ports can theoretically aggregate much more traffic than a single one-gigabit uplink can carry. In real office networks, not every port transmits at line rate simultaneously, so this oversubscription is normal. The design still needs to reflect the actual traffic profile. A site dominated by cloud applications and internet traffic may use a modest uplink comfortably, while a media-production or data-transfer environment can saturate an uplink much faster.

If the intended deployment includes modern Wi-Fi access points with multigigabit Ethernet interfaces, the non-X MS130-48P deserves particular scrutiny. The switch provides 1 GbE access ports rather than 2.5 GbE mGig ports. An access point can often operate on a gigabit connection, but doing so may constrain its potential aggregate wired throughput. Where multigigabit access is part of the wireless design objective, compare the MS130-48X or another appropriate model instead of selecting the MS130-48P only because the PoE budget looks attractive.

PoE planning: the 740 W budget is powerful, but it still needs arithmetic

A practical PoE worksheet usually lists each device class, quantity, expected maximum power, and a reserve factor. Suppose a site has twenty access points, twenty phones and eight cameras. The correct calculation is not simply forty-eight powered ports. The design team should check the maximum draw of the exact AP, phone and camera models, then compare the total with both the 740 W chassis budget and the 30 W per-port capability. If each AP were a relatively demanding device while the phones and cameras were light loads, the result might still fit comfortably. If the cameras include heaters, infrared illuminators or other high-consumption functions, the design could look very different.

This arithmetic also affects resilience. During a utility outage, the switch and every PoE endpoint connected to it draw from the UPS. A large switch with hundreds of watts of powered devices can exhaust a small UPS quickly even though the network remains logically healthy. Buyers should therefore treat PoE budget and backup-power runtime as one planning exercise rather than two unrelated procurement items.

Uplink design: understand the four 1 GbE SFP ports before ordering optics

The MS130-48P provides four 1 GbE SFP uplink ports. This is a deliberate distinction from the MS130-48X, which provides faster SFP+ uplinks. For many traditional branch and campus access layers, one-gigabit fibre uplinks remain sufficient, especially where application traffic is internet-bound, user concurrency is moderate and multiple switch uplinks are distributed across the architecture. For bandwidth-intensive sites, however, the uplink choice may become the first design constraint.

Supported Cisco Meraki SFP options for the MS130-48P include MA-SFP-1GB-SX, MA-SFP-1GB-LX10 and MA-SFP-1GB-TX. The correct module depends on media and distance. SX is used for appropriate multimode fibre applications, LX10 for suitable single-mode fibre applications, and the copper transceiver option is relevant where a supported RJ45 uplink is needed. The transceiver should be matched to the opposite-end interface, fibre type, connector path and optical budget. Ordering “an SFP” without those details is not enough for an accurate deployment.

A second consideration is how many uplinks are actually needed. Some deployments use a single uplink, some use link aggregation for additional capacity or resilience, and some connect to separate infrastructure according to a broader redundancy design. The MS130-48P’s four SFP ports provide physical flexibility, but the surrounding network determines whether that flexibility can be used effectively. The upstream switch must support the intended link design, VLANs and any aggregation configuration.

When evaluating the uplink requirement, estimate traffic generated by the connected edge population rather than looking only at access-port quantity. Forty-eight desk phones and ordinary office computers can have a very different traffic profile from forty-eight high-resolution cameras or a dense collection of access points. If sustained aggregate traffic can exceed the intended upstream capacity, choose an architecture with faster uplinks or distribute the load appropriately instead of assuming the switch fabric alone will solve the bottleneck.

Meraki Dashboard management and operational value

The MS130 family is managed through the Cisco Meraki Dashboard, which is central to the product’s value proposition. A new switch is normally claimed into a Dashboard organisation, added to a Dashboard network, physically connected, allowed to check in, and then configured through the cloud management interface. This approach can reduce the amount of device-by-device command-line work required at distributed sites and can make standardised configuration easier to maintain across a fleet.

Remote operational tools are particularly useful when the IT team does not have engineers permanently present at every branch. Meraki documentation highlights remote packet capture capabilities, automatic firmware upgrades, SNMP and syslog integration, and central network visibility. A support team can use the Dashboard to investigate port states and network events without beginning every incident with a site visit. That can improve response time, but organisations should still define escalation paths for issues involving cabling, physical power, damaged endpoints or other conditions that cannot be repaired remotely.

Cloud management also changes how change control should be handled. Configuration is no longer tied to one local console session. Administrators should use suitable access-control practices for the Dashboard organisation, define who can change switch settings, document VLAN and port-profile standards, and coordinate firmware maintenance with business requirements. The simplicity of a cloud interface should not be mistaken for permission to manage a production access layer casually.

The MS130 supports common access-layer functions such as 802.1Q VLAN tagging, 802.1X authentication, DHCP snooping, IPv4/IPv6 ACL support and DHCP relay. These capabilities allow the switch to participate in a properly segmented branch design rather than operating as a flat unmanaged Ethernet concentrator. The exact policy architecture should be aligned with the organisation’s identity platform, DHCP design, firewall segmentation and wireless environment.

For businesses already invested in the Meraki ecosystem, the centralised operational model may be the decisive factor. For businesses using another switching platform with established automation, telemetry or operations processes, the value should be judged in the context of that existing environment. A technically capable switch can still create unnecessary operational fragmentation if it introduces a new management and licensing system for a very small number of devices.

Licensing is a purchase dependency, not an optional accessory

Cisco Meraki licensing needs to be planned at the same time as the MS130-48P hardware. The MS130 family supports Enterprise and Advanced licence tiers in Meraki’s classic licensing structure, with specific 48-port licence SKUs. Cisco documentation lists LIC-MS130-48-xY for the 48-port Enterprise licence and LIC-MS130-48A-xY for the 48-port Advanced licence, where the term is reflected in the SKU. Enterprise is available in 1, 3, 5, 7 and 10 year terms, while the MS130 Advanced licence is not offered in the 7 and 10 year terms in the same way.

The organisation’s licensing mode can affect what combinations are valid. In co-termination licensing, an organisation that uses MS130 switches must keep the relevant supported switch licensing tier aligned rather than casually mixing Enterprise and Advanced on applicable models. Organisations that already contain models such as MS390, C9300-M or MS150 with a particular tier need to check how the addition of an MS130 affects consistency. Under per-device licensing, tier combinations can differ, but certain features may still have organisation-wide requirements.

Subscription licensing is also available for the MS130 family. The MS130-48P falls into the MS100 Large subscription grouping. Cisco lists Essentials and Advantage variants for that class. The correct commercial model depends on how the customer currently licenses Meraki infrastructure, the desired contract structure and whether the organisation is migrating to subscription licensing. For an existing Meraki customer, matching the current licensing architecture is usually more important than choosing a term in isolation.

The Advanced tier should not be treated as a way to change the hardware characteristics of the MS130-48P. A licence does not convert its 1 GbE SFP uplinks into 10 GbE ports, nor does it add 2.5 GbE mGig access interfaces. Cisco’s MS130 documentation associates Adaptive Policy readiness specifically with relevant hardware models and firmware conditions, so any advanced-feature requirement must be verified against the exact model rather than inferred from the existence of an Advanced licence SKU.

An accurate quotation therefore needs the hardware quantity, current Meraki organisation status, licensing model, desired term and feature requirements. Omitting licence planning can produce an incomplete commercial comparison because a lower hardware price does not represent the full operating requirement.

Security and segmentation at the access layer

An access switch is where users and devices physically enter the wired network, so it plays a practical role in segmentation and access control even when the firewall remains the main security enforcement point between zones. The MS130 supports 802.1Q VLAN tagging, which allows different groups of endpoints to be separated logically across the same physical switching infrastructure. Typical designs may use different VLANs for corporate users, voice, cameras, building systems, guests or other device classes, depending on the organisation’s security policy.

802.1X authentication can be used where the organisation has the necessary identity and RADIUS infrastructure to validate network access. That capability is useful for businesses trying to move beyond simple port-based trust, but the switch alone does not create an identity architecture. Certificates, supplicant behaviour, authentication policy, fallback methods and treatment of devices that do not support 802.1X all need planning. A rushed rollout can create user disruption even when the switch supports the desired feature correctly.

DHCP snooping can help protect the access layer from unauthorised DHCP behaviour when configured in line with the network design. IPv4 and IPv6 ACL support provides another mechanism for enforcing permitted traffic patterns at the switching layer. These features should complement, not randomly duplicate, firewall policy. A clear division of responsibility between access control, inter-VLAN routing, firewall inspection and endpoint security makes troubleshooting much easier.

The strongest security outcome usually comes from accurate device classification, sensible VLAN boundaries, authenticated access where appropriate, protected management access and consistent monitoring. Purchasing a cloud-managed switch is only the first step. The configuration model determines whether the hardware is being used as a secure access-layer platform or merely as a remotely managed collection of Ethernet ports.

Deployment fit by environment

Dubai and UAE installation considerations

The MS130-48P is specified for operation from 0°C to 45°C. In Dubai and across the UAE, that specification makes the condition of the communications room important. A correctly air-conditioned indoor rack can remain well within the required range, while an inadequately cooled cabinet, warehouse enclosure or utility room can exceed acceptable temperatures even when the surrounding office feels comfortable. Switch selection should therefore be accompanied by a review of ventilation, rack loading and cooling reliability.

Power planning is equally important because the switch can support a large powered-device population. The maximum documented power load is far above the idle figure, so electrical circuits, PDUs and UPS systems should be sized for the anticipated PoE load and desired backup runtime. If the same UPS also powers a firewall, router, optical network equipment, servers or other switches, its total load should be modelled as a system rather than device by device.

The switch is 1U high and approximately 34 cm deep. That depth is manageable in conventional enterprise racks, but shallow wall-mounted cabinets should be checked before delivery. Patch-panel depth, front cable-management space, PDU placement and bend radius for fibre patch cords all affect whether an installation remains serviceable. A rack that can physically hold a 1U chassis is not necessarily deep enough for clean cabling and airflow.

Cisco Meraki documentation notes that region-specific power cords are generally ordered separately for these full-size MS130 models. For a UAE deployment, the quotation should therefore include the appropriate local power-cord requirement rather than assuming the box contains the necessary cord. Rack-mount screws are included for the full-size models, but optics, fibre patch leads, copper patch cords and other deployment materials depend on the site design.

For projects involving several branches, a staged deployment plan can reduce risk. Claiming and pre-configuring devices in the Dashboard before site installation allows VLANs, port policies and management settings to be prepared centrally. The installation team can then focus on physical rack placement, uplinks, endpoint patching and validation. Pre-provisioning does not remove the need for a tested rollback plan when replacing an existing production switch.

Migration from an existing 48-port switch

A switch replacement is often presented as a straightforward port-for-port exercise, but a reliable migration needs more detail. Before disconnecting the existing switch, export or document the current VLAN assignments, trunk settings, voice VLAN behaviour, port descriptions, authentication settings, link aggregation, spanning-tree dependencies, DHCP protection features and any static endpoint requirements. A forty-eight-port chassis can have years of historical configuration that is not obvious from the patch panel.

Physical port mapping should be prepared before the change window. Label each patch lead or map the existing patch-panel positions to the new switch ports. This is especially important when the old switch has been patched organically over time. Reconnecting cables in arbitrary order can work electrically but break assumptions embedded in port-based VLANs, voice configuration or access-control policies. Meraki port profiles and central configuration can simplify future standardisation, but the first migration still needs accurate source information.

PoE migration requires special attention because endpoint behaviour may change when power is removed and restored. Phones reboot, access points rejoin controllers or cloud platforms, cameras restart and building devices may take time to reconnect. The change plan should therefore distinguish network convergence from endpoint service recovery. A switch may be fully online while dependent systems are still booting or renegotiating power.

Uplinks should be pre-validated with the correct SFP modules and fibre path. If the existing switch uses 10 GbE uplinks, replacing it with an MS130-48P would be a downgrade in upstream interface speed. That is exactly the kind of architectural mismatch that should be discovered during design rather than during installation. If the existing uplink is 1 GbE fibre, the MS130-48P may be much more directly aligned.

After cutover, validation should include Dashboard connectivity, uplink stability, VLAN reachability, endpoint authentication, DHCP operation, PoE status, voice calling, wireless access-point health, camera connectivity and monitoring integration. A migration is complete when the business services attached to the switch are working as expected, not merely when the status LED is green.

MS130-48P versus nearby Meraki options

The comparison shows why the “P” and “X” distinctions matter. The MS130-48P is not simply the premium version of the MS130-48, and the MS130-48X is not simply a faster replacement in every project. The P model keeps all forty-eight access ports at 1 GbE while adding the large PoE budget. The X model sacrifices some pure 1 GbE port count to introduce eight 2.5 GbE mGig interfaces and uses 10 GbE SFP+ uplinks. The correct choice depends on endpoint and uplink requirements.

A buyer with forty-five ordinary wired endpoints and twenty-five PoE phones may prefer the MS130-48P because nearly every copper port can be used at one gigabit and the PoE budget is ample. A buyer building a new Wi-Fi environment with several multigigabit APs and heavy upstream traffic may prefer the MS130-48X even if not every port is used. Selecting by list price or port count without matching the traffic model can create a false economy.

Accessories and quotation dependencies

The hardware unit is only one line item in a complete deployment. Cisco’s documentation lists supported one-gigabit SFP modules for the MS130-48P, and the correct selection depends on whether the uplink uses multimode fibre, single-mode fibre or copper. Fibre patch cords must match connector type and polarity, while existing fibre runs should be confirmed for distance and condition. If a buyer requests four SFP modules but only one uplink is required, that may add cost without operational benefit. If two redundant uplinks are planned, the quantity should reflect that design.

Power-cord selection should be confirmed for the UAE installation. Cisco indicates that region-specific power cords are not generally included with the relevant full-size MS130 models outside the US automatic inclusion rule. The quotation should therefore identify the required cord rather than assuming a generic lead will be available on site. This is a small line item, but missing it can delay an otherwise complete installation.

Rack and cabling materials may include patch leads, cable managers, labels and potentially additional rack power equipment. A switch with forty-eight copper ports can create substantial cable density at the front of a rack, so horizontal cable management and sensible patch-cord lengths improve serviceability. Very long patch leads folded into a shallow cabinet can obstruct airflow and make future fault isolation difficult.

Licensing should be quoted alongside hardware with the correct term and tier. If the organisation already uses Meraki, the supplier should know the existing licensing model before recommending a licence SKU. New deployments need a decision on whether the commercial model will be classic licensing or subscription licensing, subject to current Cisco programme availability and the customer’s procurement preference.

Installation and configuration services are separate decisions. Some customers want supply only; others want pre-staging, rack installation, VLAN configuration, endpoint migration, testing, documentation and post-change support. A quotation is more accurate when that scope is stated explicitly. FourTeck IT Services UAE can be used as a reference point for broader implementation and support requirements.

When the MS130-48P may not be the best choice

The MS130-48P is not ideal when the network specifically requires 10 GbE uplinks from the access switch. Its four uplinks are 1 GbE SFP, so an organisation expecting heavy aggregate traffic should not rely on the 104 Gbps switching capacity as proof that the upstream path is equally fast. Switching capacity describes internal forwarding capability across the designed interfaces; the practical path to the rest of the network remains bounded by the chosen uplinks.

It may also be a weak fit when several connected devices need 2.5 GbE Ethernet. This situation is increasingly relevant to newer wireless access points. The MS130-48P’s copper ports are gigabit ports. If multigigabit edge performance is a project objective, the MS130-48X or another suitable access model should be assessed before purchase. A higher PoE budget does not compensate for an access-speed mismatch.

The model is also unnecessary when PoE is not required. The MS130-48 provides the same basic 48 x 1 GbE plus 4 x 1 GbE SFP architecture without the PoE subsystem. Organisations connecting only workstations, printers or other self-powered equipment may prefer to avoid paying for unused PoE capability, provided future plans do not introduce powered endpoints.

A forty-eight-port chassis can be excessive for a small branch with fewer than twenty active Ethernet devices and little growth. In that case, a smaller port-count model can reduce cost and power consumption. Conversely, a very large site should consider whether one forty-eight-port access switch creates a single operational or power-domain concentration that is undesirable. Multiple switches can provide more capacity and layout flexibility, though they also introduce additional hardware and licence requirements.

Finally, the Meraki cloud-management model should fit the organisation’s operational strategy. A company standardised on another vendor’s campus architecture may need a stronger reason than hardware specifications alone to introduce a separate management and licensing platform. The right switch is the one that fits both the technical requirements and the operating model.

Practical design scenarios

Operational checklist after installation

Buying the Cisco Meraki MS130-48P in Dubai

A useful Dubai quotation should describe more than “one Cisco switch.” It should identify the exact MS130-48P hardware, required quantity, licence type and term, region-appropriate power requirement, SFP optics, patching, implementation scope and support expectations. If the customer is replacing an existing switch, the quotation may also need migration services and an agreed change window. If the customer is expanding an existing Meraki organisation, the licensing mode should be reviewed before the commercial offer is finalised.

Availability and commercial terms can change, so stock and delivery should be confirmed at quotation time rather than represented as permanently fixed. For urgent projects, it can be useful to identify whether a functionally suitable nearby model is acceptable if the exact MS130-48P delivery lead time does not meet the project schedule. Substitution should be based on requirements, not simply on what is available; a 48X, for example, changes both access-port composition and uplink capability.

FourTeck can support projects that combine switching with wider network-security and infrastructure requirements. Buyers comparing access switching alongside firewalls or branch connectivity can review Firewall Dubai by FourTeck. Organisations with regional or broader procurement requirements can also use FourTeck global as an additional corporate reference.

For the most accurate offer, provide the current network topology and the outcome you are trying to achieve. A short requirements discussion can often reveal whether the MS130-48P is the right model, whether the non-P version is sufficient, or whether the 48X is more appropriate because of multigigabit access and 10 GbE uplinks. That comparison is more useful than simply confirming that the requested SKU can be supplied.

Frequently asked buyer questions

Support, lifecycle and ongoing administration

A switch remains part of the production network for years, so the purchase decision should include the support model after installation. Meraki’s cloud-first approach simplifies several recurring tasks, including firmware management and remote visibility, but an organisation still needs someone responsible for change control, incident response, licence renewal and configuration governance. Whether that responsibility sits with internal IT, a managed service provider or a hybrid team should be clear before the network expands.

Firmware updates should be treated as planned operational events. Automatic or centrally scheduled update capabilities reduce manual work, but change windows should still reflect the business impact of an access-switch restart. A branch switch can affect phones, wireless access, cameras and users at the same time. Larger estates may benefit from staged rollout policies so a new firmware release is validated at selected sites before broad deployment.

Licence renewal is another lifecycle dependency. The responsible team should record renewal dates, contract details and the relationship between hardware and the customer’s Meraki organisation. Procurement records should include the exact device model and licence model so future renewals do not begin with uncertainty about the original purchase. Subscription and classic licensing each have their own operational workflow, so the internal asset register should reflect which model is being used.

Physical maintenance should not be ignored simply because management is cloud based. Dust accumulation, damaged patch leads, stressed fibre, blocked ventilation, unstable power and failing UPS batteries can all affect network reliability. Regular rack inspections and clear cable labelling make remote troubleshooting more effective because a remote engineer can give precise instructions to local staff when physical action is required.

Lifecycle planning should also include a capacity review. Port utilisation, PoE consumption and uplink utilisation can indicate whether the original design still fits. If the site begins deploying multigigabit access points or consistently approaches one-gigabit upstream limits, that operational data can support a planned upgrade rather than an emergency replacement.

Procurement guidance for larger rollouts

When several MS130-48P switches are being purchased for a campus or multi-branch project, standardisation creates both benefits and constraints. A common switch model simplifies spares, training and templates, but it can be inefficient if different sites have materially different requirements. A twenty-user branch may not need a forty-eight-port PoE switch, while a high-density wireless site may need multigigabit ports and faster uplinks. Standardise around validated site profiles rather than forcing one model into every cabinet.

A bill of materials should separate hardware, licences, optics, power cords and services. This makes commercial comparisons more transparent and reduces the risk of an apparently low quote that omits necessary items. For fibre-connected branches, identify transceiver type and quantity by site. For PoE-heavy sites, include UPS capacity review. For migration projects, distinguish pre-configuration, onsite cutover and after-hours support rather than bundling all labour into an undefined line.

Serial-number and asset tracking should be considered before deployment starts. Devices can be pre-assigned to locations, rack positions and Dashboard networks so implementation teams know exactly where each unit belongs. This discipline becomes more valuable as rollout quantity increases because accidental installation of the wrong device at the wrong site can complicate inventory and support records.

For multi-country organisations, commercial and regulatory requirements may differ by destination. A UAE project should use the appropriate local power and procurement requirements, while another country may need different logistics. FourTeck’s broader regional presence can be referenced through FourTeck Africa where projects extend into African markets, while UAE supply and implementation can remain coordinated locally.

The technical standard should also document exceptions. If some sites require MS130-48X because of 10 GbE uplinks, or smaller MS130 models because of port count, record why. This turns product selection into an auditable architecture rather than a collection of ad hoc purchasing decisions.

Decision recap: six points that determine whether the MS130-48P fits

What FourTeck needs from you for an accurate quotation

Build the right MS130-48P configuration for your Dubai network

If your requirement is a 48-port cloud-managed PoE access switch, the MS130-48P can be a strong fit when gigabit access, 1 GbE SFP uplinks and a 740 W PoE budget match the design. The next step is to confirm licences, optics, endpoint power, rack conditions and migration scope so the quotation reflects a deployable solution rather than hardware alone.