Are You Ready for 40G and 100G?

Data centers regularly undertake their own great migration, to ever higher speed networks. 10G, unimaginable a decade ago, is now common in larger enterprises. And now many enterprises have to adopt 40 Gigabit Ethernet or even 100G in the aggregation and core layers of data center networks in order to meet the overall bandwidth demands of top-of-rack servers. The need is clear: a 40/100G Ethernet migration plan is quickly becoming a matter of survival. Is your network cabling optimized for this inevitable growth? Are you ready for 40G and 100G?

Fiber Transmissions at Higher Speeds

When moving to 40/100GbE, the most important difference in backbone and horizontal multimode applications is the number of fiber strands. 40GBASE-SR4 uses 8 strands in total, 4 strands to transmit and 4 to receive. 100GBASE-SR10 uses 10 lanes to transmit and another 10 lanes to receive for a total of 20 strands. In data centers and backbones, it may be possible to have 8 or 20 individual strands of fiber. However, those strands may take disparate paths from one end to the other and this can cause delay skew (known as bit skew) resulting in bit errors. For this reason, the 40/100GbE standards are written around fiber optic trunk assemblies that utilize a MPO/MTP multi-fiber array connector. Data is transmitted and received simultaneously on MTP interfaces through 10G simplex transmission over each individual strand of the array cable. In these assemblies, all strands are the same length. Also referred to as "parallel optics", this construction minimizes bit/delay skew, allowing the receive modules to receive each fibers information at virtually the same time.

Copper Transmissions at Higher Speeds

The first 10GbE capable copper interface was developed for the 10GBASE-CX4 application. The physical requirements for this shielded four-lane copper connector is standardized under SFF-8470. As a passive assembly, the SFF-8470/CX4 cables have a reach of 15m. This assembly supports 10GbE, InfiniBand, FibreChannel and FCoE. These assemblies use twinax cable, constructed of two inner conductors with an overall foil covered by a braid shield. Due to their low latency, these cables are popular in supercomputing clusters, High Performance Computing and storage. As part of the 802.3ba 40/100GbE standard, multi-lane 40GBASE-CR4 and 100GBASE-CR10 was defined. This standard specifies the use of 4 and 10-lane twinax assemblies to achieve 40 and 100GbE speeds for distances up to 7m.

40/100G MPO/MTP System

MPO/MTP is available in both 12 and 24 strand termination configurations used at the end of a trunk assembly. A modular laser optimized multimode MPO/MTP system that supports 40G and 100G fiber optic networks includes trunks, harnesses, array cords, modules, and adapter plates. For 40GbE, a 12-fiber cabling solution with each channel featuring four dedicated transmit fibers and four dedicated receiver fibers is used. In general, the middle four fiber remain unused. Parallel transmission is also used for 100GbE with a 24-fiber solution or two 12-fiber solution.

At present, 40GbE is taking over from 10GbE as the new high-growth market segment. Meanwhile, the 40GbE optics are universal in data center and the market of 100GbE is accelerating. Being prepared for 40/100G is essential: within a few short years higher-speed Ethernet will be common in data centers across all types of organizations. Fiberstore is ready. Fiberstore offers various optical communication products to meet diverse demands. For example, we provide 40GBASE-SR4 QSFP+ transceivers, like Finisar FTL410QE2C 40GBASE-SR4 QSFP+ transceiver and Mellanox MC2210411-SR4 40GBASE-SR4 QSFP+ transceiver, which are branded by famous companies and quality guaranteed.

Upgrading to 40/100 Gigabit Ethernet

To support the changing and fast-growing bandwidth demands of data centers (DCs), the IEEE ratified standards for supporting 40 Gigabit Ethernet (40GE) and 100 Gigabit Ethernet (100GE), known as IEEE 802.3ba. 40GE and 100 GE belong to a family of network technologies that primarily are used in local area networks. As they are becoming more and more popular, it is very necessary to know more about them. This post offers a quick-reference guide to increase your familiarity with issues in the migration to 40 and 100GE systems.

Cabling Systems

Both 40 and 100 Gigabit Ethernet can be deployed using the same cabling systems in use today. Multi-mode will employ parallel optics using MPO interconnects and require additional cable infrastructure depending on the system deployed while single mode fiber will employ serial transmission and use LC or SC connectors. The approach used for the higher speed data rates is based on advanced transceiver technologies engineered to take advantage of the full bandwidth of laser optimized fibers. Both single mode fiber (SMF) and multi-mode fiber (OM3, OM4) were approved in the standard.

Standards

• 40GBase-SR4: multimode fiber with four fiber pairs (OM3/OM4)
• 40GBase-LR4: single-mode fiber pairs with 4 wavelengths a 10Gbit/s at 10 km
• 40GBase-KR4: short distances in the backplane (4 pairs of wires)
• 40GBase-CR4: twinax cable with 4 pairs
• 100GBase-CR10: twinax cable with 10 pairs
• 100GBase-SR10: multimode fiber with 10 fiber pairs (OM3/OM4)
• 100GBase-SR10: single-mode fiber pairs with 4 wavelengths a 25Gbit/s at 10 km
• 100GBase-ER4: single-mode fiber pairs with 4 wavelengths a 25Gbit/s at 40 km

Module Types

QSFP: The QSFP (quad small-form-factor pluggable) is the primary way of delivering 40GE and are now appearing in 100GbE-capable form. These present either MPO connectors for multi-mode optics or LC connectors for single-mode optics. It is similar in size to the CXP and provides four transmit and four receive lanes to support 40GbE applications. QSFP+ (quad small form-factor pluggable plus), as the updated version of QSFP, is a new multi-source agreement (MSA) for high speed application. QSFP+ gradually replace QSFP and is widely used by people as it can provide higher bandwidth. The following picture shows an Avaya AA1404001-E6 compatible QSFP+ transceiver.

CFP: The CFP (C form-factor pluggable) transceiver features twelve transmit and twelve receive 10Gb/s lanes to support one 100GbE port, or up to three 40GbE ports. CFP form factor optics are available in 40GbE and 100GbE varieties. Its larger size is suitable for the needs of single-mode optics and can easily serve multi-mode optics or copper as well. These present MPO connectors for multi-mode optics or SC connectors for single-mode optics. CFP modules use the 10-lane CAUI-10 electrical interface.

CFP2: The CFP2 is an evolution of the existing CFP form factor, using manufacturing and optics advances to reduce the size of the module to approximately half that of the original CFP, allowing for higher interface density. CFP2 modules use the 10-lane CAUI-10 electrical interface or the 4-lane CAUI-4 electrical interface.

Requirements and Objectives

40GE and 100GE fulfill the following requirements and objectives:

• Preserve existing 802.3 frame format, minimum size, and maximum size.
• Support high-speed switching, routing, and application functions in data centers.
• Exhibit a bit error rate (BER) of 10-12 or better.
• Provide support for optical transport network (OTN).
• Provide specifications for operation over single-mode optical fiber, laser optimized multimode optical fiber, copper cables, and backplanes.
• Support high-bandwidth applications such as video on demand (VoD) and high-performance computing (HPC).

The throughput and reliability of a 40/100 Gigabit Ethernet installation is directly related to the cabling, the choice of connectivity, and proper installation. Fiberstore's superior cable and connectivity designs provides you with superior levels of performance and operation. We offer all kinds of cables and transceiver modules for your applications, such as Dell Force10 GP-QSFP-40GE-1SR 40GBASE-SR4 QSFP+ transceiver, or Juniper QFX-QSFP-DAC-3M compatible QSFP+ to QSFP+ passive copper cable, and so on. You can buy from us with confidence.