Direct answer: what are Juniper 100G optics?

Juniper 100G optics are pluggable optical transceivers and related high-speed interconnect options designed to carry 100 Gigabit Ethernet across links ranging from short data-center connections to metro, WAN and long-distance transport, depending on the exact optic. Juniper’s current 100G portfolio spans direct-detect and coherent technologies, and commonly includes QSFP28 modules as well as other form factors for particular generations of hardware.

They are mainly used to connect Juniper switches, routers and security platforms to other network devices over multimode or single-mode fibre. Organizations should consider them when a 100G port must be activated, a 40G or lower-speed uplink is being upgraded, a spine-leaf fabric is expanding, a router-to-router link needs greater capacity, or a service-provider handoff requires a supported 100GbE optical interface.

The most important factor to confirm is compatibility of the exact transceiver with the exact Juniper host platform, interface module and software release. A 100G QSFP28 mechanical fit does not by itself prove operational support. Fibre type, connector style, optical reach, breakout mode, FEC expectations, temperature grade and the far-end optic must also match the intended link.

FourTeck can help determine the appropriate Juniper-qualified optic for the platform, required reach and installed fibre, then identify the patch cords, MPO/MTP or LC connectivity, breakout cables and quotation details needed for deployment in Dubai or elsewhere in the UAE.

Why 100G optics are a selection problem, not just a speed purchase

A request for “Juniper 100G optics” sounds simple until the physical link is examined. Two 100G transceivers can share the same headline Ethernet speed while using completely different fibre plant, connectors, wavelengths, lane architectures and reaches. One module may be designed for parallel multimode fibre across a data hall. Another may carry the same 100GbE service over duplex single-mode fibre for several kilometres. A third may use single-lane PAM4 signaling with FEC expectations that must be understood at the host interface. A coherent optic may be selected for a transport-oriented application that is materially different from ordinary client optics.

This is why part-number-level selection matters. Juniper publishes platform qualification through its Hardware Compatibility Tool and product documentation. Compatibility can be tied not only to the chassis or fixed switch model, but to the specific line card, PIC, FPC or port type and to the Junos OS release that introduced support. A module that is supported in one 100G port on one platform should not be assumed to be supported in every port that accepts a QSFP-shaped transceiver.

For Dubai buyers, a clean procurement process starts with the link design. Identify the two endpoints, distance, fibre medium, connector presentation, speed mode and whether breakout is required. Then map those conditions to a qualified optic. This order of work reduces the risk of receiving perfectly genuine hardware that still cannot be used in the intended interface.

Core Juniper 100G optic types buyers commonly evaluate

Quick selection logic for a Dubai 100G fibre link

Begin with the installed fibre and distance, then validate the chosen technology on the exact Juniper platform. The table below is a planning shortcut, not a substitute for platform qualification.

QSFP28 is common at 100G, but form factor is only the first gate

QSFP28 became a major 100G form factor because it can carry four 25G electrical lanes and provide a compact pluggable package. Juniper’s documentation also distinguishes other 100G form factors including CFP and CFP2 for legacy or particular line-card designs, plus newer compact architectures used on certain platforms. The practical lesson is that the transceiver slot and host electrical interface must be identified before a module is ordered.

A QSFP28 module cannot be inserted into a CFP or CFP2 port, and the reverse is also true because these are physically different formats. Even where a port accepts QSFP28, the port may need to be configured for the intended data rate and the platform may support only a defined subset of optics. Juniper notes that some QSFP28 ports can be used with QSFP+ modules at 40G when the port is configured appropriately; that capability should never be generalized to every platform without checking the hardware documentation.

For a buyer, this means the phrase “100G QSFP” is incomplete. A useful purchase request names the platform model, port or interface card, intended 100GbE mode, fibre type, approximate distance and far-end device. Those six items usually eliminate most unsuitable choices before detailed optical-budget work begins.

Understanding 100G breakout before you buy

Breakout allows a high-speed physical port or optical assembly to provide multiple lower-speed logical links, but not every 100G optic supports it. Juniper identifies JNP-QSFP-100G-SR4 as breakout capable and lists MTP-to-four-LC breakout cable options for 4 × 25GbE use. JNP-QSFP-100G-PSM4 is also documented as breakout capable with appropriate MTP-to-four-LC single-mode breakout cabling. By contrast, Juniper marks common CWDM4, DR, FR and LR/LR4 optical modules as non-breakout in their respective Hardware Compatibility Tool entries.

This matters operationally because a customer may intend to use a 100G switch port as four server-facing 25G connections. Buying a 100G duplex optic simply because it fits the port will not create four independent 25G optical paths. Breakout requires a supported port mode, compatible transceiver or cable, suitable fan-out assembly, correct interface configuration and compatible 25G optics or ports at the far end.

When a quote includes breakout, specify the number of 25G endpoints, required cable length, fibre type and the exact Juniper port. It is also wise to identify whether the fan-out will terminate directly on equipment or through a patching system, because cassette polarity and connector gender can become part of the design.

Compatibility is a four-layer check

Juniper platform families and why the exact host matters

Juniper positions its 100G optical portfolio across routing, switching and other high-performance networking environments, and individual optics appear as qualified options on numerous ACX, EX, MX, PTX, QFX and selected SRX platforms. This does not mean one universal 100G optic is supported everywhere. Hardware generations use different ASIC interfaces, line-card designs, thermal envelopes and software qualification matrices.

On a fixed switch, a buyer may only need the exact switch model and port number. On a modular router, the chassis name alone can be inadequate because the optical interface belongs to a particular MPC, line card, PIC or FPC. An optic supported on one card in an MX chassis may not be supported on another card in the same chassis. The same concept applies when a platform has multi-rate ports with different operating modes.

The software release also matters. Juniper’s compatibility records frequently show an “Introduced Release” for a transceiver on a particular platform. If a production network is intentionally held on an older Junos release, a newly added optic might require a software decision rather than a simple hardware swap. That creates a change-management issue, not just a procurement issue.

For quotation accuracy, provide a photo or inventory output showing the host model, interface card and current Junos version whenever these details are uncertain. That small amount of information can prevent an incompatible optics order and can reveal whether a code upgrade should be planned before installation.

Fibre medium: multimode versus single-mode at 100G

The installed fibre often narrows the optic choice more quickly than any other factor. SR4 is a parallel multimode technology intended for comparatively short links. PSM4 uses parallel single-mode fibre and extends beyond typical SR4 distances while preserving a multi-fibre presentation. CWDM4, DR, FR, LR and LR4 families use single-mode fibre with duplex LC connectivity in the common variants discussed on this page.

Do not identify fibre by jacket color alone. Existing installations should be verified from cable markings, patch-panel records or test documentation. Multimode categories and single-mode construction affect loss and supported distance. MPO-based systems also require attention to polarity, fibre count and connector gender. APC and UPC physical-contact finishes are not interchangeable simply because the outer connector family looks similar.

For a new data-center build, parallel MMF may be attractive where equipment rows are close and breakout flexibility is valuable. For a campus backbone or inter-building connection, duplex single-mode fibre is often operationally convenient because the same fibre plant can serve multiple optical generations and reaches. For an existing site, the best choice is frequently the qualified optic that uses the fibre already installed and tested, provided the loss budget and distance are suitable.

If the fibre route includes patch panels, cassettes, cross-connects or building-entry facilities, count all connector pairs and splices when considering the optical budget. “Cable length” alone does not describe total channel loss.

Connector planning: LC, MPO/MTP, UPC and APC

Juniper 100G optics use several connector presentations. Common SR4 modules use an MPO-12 PC/UPC interface. PSM4 modules use MPO-12 APC. Many duplex single-mode 100G optics, including CWDM4, DR, FR and LR/LR4 variants, use duplex LC PC/UPC connectors. These details should appear in the bill of materials because the wrong patch lead can stop a deployment even when the optics are otherwise correct.

MPO/MTP infrastructure deserves special care. Parallel optics use multiple transmit and receive fibres, so polarity must maintain the correct lane mapping from one end to the other. Patch cassettes and trunks can change how fibres are presented. A breakout application adds another layer because one MPO connector may fan out to multiple duplex LC pairs. The fibre type and polish must match the optic specification.

LC links are simpler physically, but cleaning and loss management are still essential. One contaminated ferrule can increase loss or cause intermittent behavior that looks like an optic fault. Inspection-before-connection and correct cleaning tools are sensible practices for new installs and troubleshooting.

FEC and signaling architecture: the hidden difference between similar-looking optics

Forward Error Correction is one of the reasons modern high-speed links can operate reliably at demanding symbol rates. Juniper’s documentation differentiates four-lane NRZ-era interfaces from newer PAM4 optical approaches. For example, the common LR4 architecture uses a legacy CAUI-4 host interface without FEC, while DR, FR and LR single-lane optical variants use media-side RS(544,514) FEC and have defined host-interface behavior. These are engineering distinctions with purchasing consequences.

A customer should therefore avoid choosing DR, FR or LR1 merely because the reach and connector appear suitable. The host must explicitly support the module and its electrical interface. At the link level, the far-end device must be compatible with the optical standard and expected FEC behavior. Problems can arise when one end is configured differently, when an intermediate device presents a different standard, or when a breakout assumption conflicts with the optic’s lane architecture.

For a straightforward Juniper-to-Juniper deployment using qualified optics, platform documentation usually provides the safest path. In a multivendor link, the standard, reach, FEC configuration and optical power specifications should be checked on both sides rather than assuming that two ports labeled “100G LR” will automatically interoperate.

Digital optical monitoring and operational visibility

Juniper lists digital optical monitoring on many 100G transceivers, including SR4, PSM4, CWDM4, DR, FR, LR and LR4 examples. Monitoring can expose values such as optical transmit and receive levels, temperature and other module diagnostics where the platform and software support those readings. This visibility is valuable during commissioning because a link can be verified against expected optical margins instead of being judged only by whether the interface comes up.

Operational teams can use DOM information to distinguish likely fibre-loss problems from configuration or hardware faults. A receive level close to the allowable boundary may indicate contamination, excessive connector loss, a damaged patch lead or an optical path that lacks sufficient margin. A healthy initial reading also creates a baseline for later troubleshooting.

DOM should not replace proper test equipment or an optical budget where the link is critical, but it is a useful part of acceptance. For new Dubai data-center and campus links, recording interface status, errors, FEC counters where applicable and optical levels at handover gives the operations team a practical reference point.

Reach is a design limit, not a buying target

An optic’s published reach is not a recommendation to use every link right at the maximum distance. The actual channel includes fibre attenuation, connector loss, splice loss, patching, ageing and measurement tolerance. An optical design needs enough margin for reliable operation under expected conditions. A 10 km-class LR optic may be selected for a 6 km campus or metro link because it fits the optical budget comfortably, not because the buyer intends to consume every kilometre of reach.

Conversely, selecting a much longer-reach optic for a very short link is not always an advantage. Some long-reach optics can require attention to receiver overload or minimum attenuation, and coherent or amplified transport systems introduce entirely different engineering considerations. The exact data sheet and platform guidance should govern the decision.

A good quotation therefore asks for route distance and, where available, an optical-loss test rather than a simple “under 10 km” statement. The farther and more critical the link, the more useful a documented loss budget becomes.

Longer-reach 100G options: when 10 km is not enough

Juniper’s 100G portfolio extends beyond the familiar SR4, CWDM4 and LR4 classes. Hardware Compatibility Tool entries include 40 km-class 4WDM-40 QSFP28 options, ER-class modules on supported hardware, and coherent 100G optics such as 100GBASE-ZR implementations for transport-oriented use cases. Juniper describes its broader 100G portfolio as covering distances from short data-center links through very long coherent applications.

These longer-reach choices should be treated as engineering projects rather than interchangeable LR replacements. A 40 km client optic may depend on a particular host interface and optical budget. Coherent optics can involve line-system, amplifier, wavelength, power and dispersion considerations that do not exist on an ordinary 2 km duplex-LC link. Some modules also have specific temperature, power and airflow requirements that matter in dense or environmentally constrained platforms.

For metro links in the UAE, the first question is whether the network needs a direct dark-fibre client optic, a service-provider Ethernet handoff, or integration with an optical transport system. Those are different architectures. If a carrier supplies the 100G service, the customer may only need a short client-side optic to the provider’s network termination. If the customer owns the fibre route, the optical reach and loss budget become the customer’s responsibility.

Provide both endpoint models, measured or estimated route loss, route distance and whether any WDM or amplification equipment sits in the path. That information determines whether an extended-reach client optic is appropriate or whether a transport solution should be evaluated instead.

DAC and AOC alternatives for very short 100G connections

Not every 100G connection needs separate optical transceivers and patch cords. Juniper also offers direct-attach copper and active optical cable assemblies for supported 100G ports. A DAC can be attractive for very short same-rack or adjacent-rack links because it integrates the cable and end modules into one assembly. An AOC provides an integrated optical path that can extend farther than passive copper while avoiding separate fibre patching between pluggable modules.

The trade-off is flexibility. Separate optics plus structured fibre allow the cable plant and transceivers to be changed independently. A DAC or AOC is a fixed assembly with a defined length, so cable-routing mistakes can require a different part. Passive DAC also has host signal-integrity limits as length increases. Juniper’s 2026 100G optics guide describes passive DAC as appropriate for very short range and notes active DAC uses additional electronics to condition the signal for longer cable lengths.

For top-of-rack switching, a DAC may be the simplest and most economical qualified option when both endpoints support the same assembly. For a structured data hall, an optical solution may provide better operational flexibility. The correct comparison should include port support, required length, cable-management constraints and whether future moves or cross-connects are likely.

Genuine Juniper-qualified optics versus third-party modules

Standards compliance can enable multivendor optical links, but platform support policy remains important. Juniper emphasizes that its qualified optics undergo system and module testing, software integration and support processes. Juniper also documents that JTAC may assist in diagnosing a problem when a third-party optic or cable is present, but the engineer can recommend checking or replacing that component with a Juniper-qualified equivalent as part of fault isolation.

For mission-critical production links, this support path is a material procurement consideration. A lower-cost third-party optic can be technically interoperable yet create an additional variable during troubleshooting. In environments where vendor support and predictable qualification are priorities, using Juniper-qualified optics can simplify the support boundary.

The commercial decision depends on risk tolerance, support policy, link criticality and lifecycle plans. FourTeck can quote Juniper-qualified parts and, when discussing alternatives, should distinguish clearly between a genuine Juniper orderable optic and any compatible third-party option rather than treating them as the same product.

Temperature, airflow and power considerations in Dubai deployments

Optical modules dissipate heat, and their allowable operating range is tied to the specific part and host platform. Several modern Juniper 100G common optics are documented with standard commercial temperature ranges, while industrial-temperature variants exist for selected families. A data-center switch in a controlled room has different environmental requirements from a router installed in a telecom shelter, outdoor cabinet or industrial location.

Dubai’s external climate does not mean a transceiver should be expected to operate at outdoor ambient temperature inside any enclosure. The relevant condition is the thermal environment specified by Juniper for the host and optic, including airflow, altitude, neighboring modules and total power density. Juniper platform documentation can impose limits on optic types or quantities under high-temperature conditions because a chassis has a finite cooling budget.

For controlled enterprise and data-center deployments, standard-temperature optics may be entirely appropriate when the room and chassis remain within specification. For an edge site or enclosure with elevated ambient conditions, confirm whether an industrial-temperature optic or a different platform is required. This is particularly important for long-reach and coherent modules, which can draw more power than simple short-reach client optics.

A reliable quote should therefore include the deployment environment for any non-standard site, not just the optic reach. If the equipment sits outside a conventional air-conditioned IT room, mention the enclosure type, cooling method and expected ambient conditions.

Installation workflow for a new Juniper 100G optical link

Migration from 10G or 40G to 100G

A 100G upgrade is often triggered by sustained uplink utilization, a new spine-leaf design, storage growth, virtualization density, internet edge expansion or the consolidation of several lower-speed circuits. The transceiver purchase is only one part of the migration. The host line card must have 100G capability, the fabric or forwarding plane must sustain the intended traffic, the far-end device must offer a compatible interface and the fibre plant must support the chosen optical technology.

Moving from 40G SR4 to 100G SR4 can sometimes preserve a parallel multimode architecture, but the exact optical budget and fibre grade still need review. Moving from a 40G LR4 duplex-SMF link to a 100G LR4 or LR1 link may be simpler physically because the duplex single-mode fibre can often be reused if the loss is within specification. A migration to DR or FR may introduce FEC and host-interface considerations that were not present on the older service.

If the old 40G port uses QSFP+ and the new port is QSFP28, remember that backward compatibility is a host feature, not an automatic optical property. Some Juniper QSFP28 ports can run 40G QSFP+ modules when configured accordingly, which can be useful during staged migration, but each target platform must be checked.

For low-risk cutover, build the new 100G path in parallel where possible, test it, then migrate routing or switching traffic during the change window. This approach allows rollback without replacing the original link under pressure.

Use cases for Juniper 100G optics in UAE networks

Procurement details that change the correct Juniper part number

The words “100G LR” are not enough for a dependable quotation. Juniper may have common optics, JNP-prefixed part families, temperature variants, equivalent common-optic SKUs and different module generations with similar marketing descriptions. A quote should preserve the exact requested SKU when the buyer already knows it. If the buyer does not know the part number, the supplier should derive it from the host and link requirements rather than selecting the first catalog entry that contains the right reach.

Quantity also affects the discussion because multi-link deployments should account for spares. A site installing dozens of optics may want a controlled spare pool, especially if the links are business-critical and a replacement cannot wait for the next shipment. The number of spare units depends on operational policy, commonality between links and local logistics rather than a universal percentage.

Cable assemblies must be quoted with equal precision. For SR4 and PSM4, state whether the requirement is a straight MPO/MTP trunk or a breakout assembly and identify connector polish and length. For duplex-LC optics, specify fibre mode and patch-cord length. For DAC and AOC, the cable length is part of the fixed assembly SKU and cannot be changed later.

If the quote is tied to a deployment date, include the required delivery window and whether installation or remote engineering support is needed. This allows stock, lead time and project services to be discussed as separate commercial items instead of being assumed.

What can make a Juniper 100G optic unsuitable?

How to compare SR4, PSM4, CWDM4, DR, FR and LR choices

The most useful comparison starts with the physical plant. SR4 is optimized for short parallel-MMF environments and supports 4 × 25G breakout on the Juniper SR4 model discussed here. PSM4 moves to parallel SMF and extends to 500 m while retaining breakout capability. CWDM4 uses duplex SMF and reaches 2 km without breakout, making it attractive when existing single-mode duplex pairs should be preserved. DR also uses duplex SMF but targets 500 m with a different single-lane optical and FEC architecture. FR extends the single-lane concept to 2 km. LR and LR4 cover 10 km-class links with different signaling generations.

No one technology is universally “best.” SR4 may be the natural answer in a data center already built around parallel MMF. CWDM4 or FR may be cleaner in a campus with duplex SMF. LR4 may be preferable where the host generation supports it broadly and a 10 km reach is required. LR1 can be attractive on newer compatible platforms. PSM4 is a specialized but useful choice where parallel single-mode infrastructure and breakout flexibility align with the design.

The real comparison is therefore an intersection of five conditions: host qualification, fibre architecture, required distance, breakout need and lifecycle plan. Cost should be compared only after those conditions are satisfied, because a cheaper module that requires replacing the fibre plant or upgrading software can have a higher project cost.

A practical 100G optics specification checklist

Lifecycle and spares planning

Optics are often purchased later than the chassis they serve, so lifecycle status deserves attention. Some Juniper 100G form factors and platform cards are mature or end-of-life, while newer systems support more recent single-lane PAM4 optics. A buyer maintaining an older installed base may need an optic that is technically correct for that generation even if a newer module exists with the same nominal reach.

For long-lived networks, standardizing on a smaller number of qualified optic families can simplify sparing, cleaning kits, patch-cord stock and troubleshooting. That does not mean forcing one optic across incompatible use cases. It means using common parts where the host and link design legitimately overlap. For example, several 2 km links on compatible platforms may benefit from a shared approved part rather than a mix of functionally similar SKUs.

When an optic is approaching lifecycle milestones, consider whether to buy spares for the expected support period or migrate the interface to a newer supported technology. The right answer depends on how long the host platform will remain in service. Buying a large spare stock for a chassis that will be replaced in six months can be wasteful; buying no spares for a critical legacy link with long lead times can create operational risk.

A lifecycle-aware quote should therefore connect optics planning to the hardware roadmap. If the 100G link is part of an upcoming refresh, mention the target platform so both current and future compatibility can be considered.

Frequently asked buyer questions

When a smaller or larger solution should be considered

100G is not automatically the right speed simply because the platform supports it. If a branch or distribution link carries modest traffic and has no near-term growth pressure, a 25G or 40G interface may meet the requirement at lower cost. Conversely, a new core design expected to exceed 100G quickly may be better served by 400G-capable hardware with 100G breakout or downspeed options, depending on architecture and budget.

The decision should consider traffic growth, port density and migration cost. Four 25G links may be operationally simpler than one 100G link in a server environment where traffic is naturally distributed. A single 100G uplink can be cleaner for aggregation between switches or routers. In a new fabric, the balance between 100G and 400G often depends on leaf-to-spine oversubscription and how long the network is expected to remain in service.

If the project is replacing equipment rather than merely activating an unused port, share the broader capacity objective. Optics should be selected as part of the architecture, not in isolation from the forwarding platform.

Quotation accuracy: information that prevents delays

A precise optics quote can be prepared quickly when the request includes the endpoint and fibre details. The biggest delays usually occur when the buyer knows the desired speed but not the platform model or installed fibre. In that situation, guessing a part number creates more risk than asking for a device inventory line, photo of the faceplate or existing cable label.

For a replacement optic, the fastest reference is often the exact SKU printed on the existing module, provided the replacement is for the same host and link. For a new link, use the platform model and physical requirements instead. If the far end is a different vendor, include that optic standard or part number so interoperability can be checked at the standard level.

If installation is required, also identify rack location, access rules, maintenance window and whether fibre testing is already complete. This separates a simple product supply request from a deployment engagement and helps avoid assumptions about patching or configuration responsibilities.

Why Juniper common optics can simplify multi-platform operations

Juniper identifies a number of optics as “Common Optic” products and maps some JNP-prefixed modules to common-optic equivalents. The operational advantage is not that every optic works in every device, but that a qualified common part can span multiple supported platform families and reduce SKU fragmentation. That can improve spare utilization when a network contains a mix of switching, routing and security systems.

Commonality should still be proven against each host. A data-center team might standardize on one supported 100G CWDM4 optic across several switch and router models, but only after verifying those models and releases in the compatibility matrix. The same principle applies to SR4, LR4 and newer DR/FR/LR common optics.

For enterprise procurement, this creates an opportunity to build a controlled approved-parts list. The list can record the Juniper SKU, supported platforms in the organization’s estate, fibre type, reach, connector and spare location. Such a reference reduces emergency purchasing errors and makes support handoffs clearer.

Commissioning and troubleshooting a 100G link

If a new 100G interface does not come up, troubleshoot in layers. First confirm the module is recognized and supported by the host. Then verify the port is configured for the intended speed or breakout mode. Check whether FEC settings match the optic and far-end requirements. Confirm the far-end interface is enabled and configured to the same Ethernet standard.

Next inspect the physical path. Confirm transmit and receive fibres are not reversed on a duplex link, or that MPO polarity is correct on a parallel link. Clean the connectors. Review receive and transmit optical levels if DOM data is available. A very low receive level points toward fibre loss, wrong wavelength technology, disconnected fibres or transmitter problems. An unexpectedly high level on a short link may indicate the need to review receiver limits for the selected optic.

When both interfaces are up but errors accumulate, examine FEC statistics, PCS errors, physical-layer counters and optical margin. A link can technically establish while operating too close to the error threshold. Compare measurements at both ends, especially on long routes. If one direction is degraded, the problem may lie in a single fibre strand, connector or transmitter.

For third-party optics, be prepared to substitute a Juniper-qualified module during diagnosis if support requests it. That substitution can distinguish a fibre or host issue from a transceiver compatibility problem.

Dubai and UAE supply considerations

For UAE projects, the main supply questions are the exact part number, quantity, delivery requirement and whether the order is tied to a planned outage or construction milestone. 100G optics are specialized components, so the correct SKU should be agreed before availability is treated as confirmed. Near-equivalent part numbers can differ in reach, temperature grade, connector or support matrix.

Large projects should identify whether all units are required at once or can be phased by site. A phased delivery may align better with installation schedules, while a single shipment can simplify acceptance and asset tracking. If optics will be stored before use, keep them in appropriate protective packaging and control dust and handling.

FourTeck can prepare a quotation around the exact Juniper 100G requirement and include related patch leads or breakout cabling where specified. Availability and lead time should be confirmed against the requested SKU and quantity at quotation stage rather than assumed from a product-family page.

Decision recap: five choices that determine the right 100G optic

What FourTeck needs from you for an accurate quotation

Send as much of the following as you have. Missing information can usually be resolved, but these inputs let the optic be selected against the real network rather than guessed from speed alone.

Specify the Juniper 100G link before you specify the optic

A correct 100G optics order ties the transceiver to the host platform, software, fibre plant, distance, connector, breakout requirement and far-end standard. Send FourTeck the endpoint and link details, and the quotation can be built around a compatible Juniper-qualified module and the cabling needed to make the connection usable—not just a part number that happens to say 100G.