DOCSLIB.ORG

The Case for 25G and 100G

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
The Case for 25G and 100G WHITE PAPER The case for 25G and 100G WHITE PAPER More bandwidth for data centers WHITE PAPER Contents Introduction ........................................................................................................................................................................ 3 Top-of-Rack Datacenter Design .......................................................................................................................... 4 25 Gigabit Ethernet (25GE) and 50 Gigabit Ethernet (50GE) Standardization ................................................................... 5 Protocol Stack ..................................................................................................................................................................... 6 25GE and 50GE switches .................................................................................................................................................... 7 25GE and 50GE NICs ........................................................................................................................................................... 8 Available Electrical and Optical Modules ............................................................................................................................ 8 25GE and 50GE Test Requirements .................................................................................................................................... 9 Xena Networks 25GE and 50GE Test Solutions .................................................................................................................. 9 Unique Eye Diagram ........................................................................................................................................ 10 Testing Above Layer 1 ...................................................................................................................................... 11 Testing from 10 Gigabit/s to 100 Gigabit/s ........................................................................................................ 12 Conclusion ........................................................................................................................................................................ 13 www.xenanetworks.com 2 WHITE PAPER INTRODUCTION For a number of years the upgrade path for higher speed Ethernet has been 10GE -> 40GE/100GE, which has been well accepted for many applications. However a couple of new speeds have recently been added between 10 Gigabit/s and 100 Gigabit/s: 25 Gigabit/s (25GE) and 50 Gigabit/s (50GE). As there an ongoing development of standards for Ethernet speeds higher than 100 Gigabit, it may seem a bit odd that lower speeds are not only discussed but in fact also standardized. The drive for 25GE and 50GE comes from datacenters, where multiple 10GE connections are used to interface to the servers in the datacenters. In high density datacenters numerous 10GE “Top of Rack” (ToR) switches are required to connect to the datacenter servers. Each switch and each switch port has a cost of its own. To this adds the cost of requirements for space, maintenance, power and cooling of the switches. By introducing 25GE single lane connections instead of the 10GE connections, the datacenters will get 2.5 times more capacity per connection, while equipment cost and space, power, maintenance and cooling requirements will be reduced compared with datacenters using only 10GE to provide the same bandwidth, simply because the datacenter with 25GE connections will be built with less equipment. In addition the CPUs in datacenter servers can now be so powerful that they easily can fill up a 10GE connection. To provide more transmission capacity the next step has until now been to upgrade to 40GE, which is considerably more expensive than a 10GE connection as 40GE is based on sending the data on 4 10 Gigabit/s lanes in parallel. As an alternative to 40GE an upgrade to 25GE will be a cost effective solution because fewer lanes are used. The general expectation in the industry is that 25GE deployment will grow rapidly in the coming years. Analysts forecast that by 2020 25GE and 50GE will together have a market share of approx. 25% in the high speed NIC market (from 10GE to 100GE) (based on data from Crehan Research). Using 25 Gigabit lanes for transport of Ethernet is not new: Most current 100GE implementations use four 25 Gigabit lanes to transport the Ethernet signals. Therefore the 25GE standardization bodies have standardized a 25 Gigabit lane based on the 100GE technology. The current standardization of 50GE is based on using two 25 Gigabit lanes. This will be an alternative to the current 40G implementations, which use four 10 Gigabit lanes to transport the Ethernet signal. The reduction of lanes will reduce the cost of the network equipment needed in the datacenter, which leads to reduced CAPEX and OPEX. In the future the Ethernet upgrade path may change to 10GE -> 25GE -> 50GE -> 100GE. In any case it will be easy and cost effective to upgrade datacenters from multiple 25GE lanes to 50GE or 100GE networks. www.xenanetworks.com 3 WHITE PAPER Figure 1: Current and future high speed Ethernet upgrade path Top-of-Rack Datacenter Design A popular datacenter design is the ”Top-of-Rack” (ToR) design: Each server rack is equipped with one Ethernet switch (or two for redundancy). The switch could be placed in the top of the rack, which has given the design its name; the switch may however be placed elsewhere in the rack. Another design is the “End-of-Rack” design, where one Ethernet switch, placed outside the server racks, connects to all the servers in the datacenter. In both cases with the increased need for bandwidth you will either need more switch ports or more bandwidth per port. Going from 10 GE ports to 25 GE ports will provide the increased bandwidth in a more cost effectively compared with adding 10 GE ports. www.xenanetworks.com 4 WHITE PAPER Figure 2: Top-of-Rack datacenter design 25 GIGABIT ETHERNET (25GE) AND 50 GIGABIT ETHERNET (50GE) STANDARDIZATION 25 Gigabit Ethernet (25GE) is a new standard for Ethernet connectivity in a datacenter environment, developed by the IEEE 802.3by 25 Gb/s Ethernet Task Force. The work was started in the IEEE standards body with a successful Call for Interest (CFI) in July, 2014. In the CFI the 25 Gigabit Ethernet Motivation was outlined: • Provide cost optimized server capability beyond 10G • Provide a 25Gbit/s MAC rate that: o Leverages single-lane 25Gbit/s physical layer technology developed to support 100GE o Maximize efficiency of server to access switch interconnect The IEEE 802.3by standard is based on technology defined for 100 Gigabit Ethernet implemented as four 25-Gbit/s lanes. As the main application of 25GE are connections inside a datacenter and inside a server rack in the data center the IEEE 802.3by standard focus on very short range connections. The IEEE 802.3by standard objectives (approved November 2014) included: • Support a MAC data rate of 25 Gb/s • Define a single-lane 25 Gb/s PHY for operation over a printed circuit board backplane • Define a single-lane 25 Gb/s PHY for operation over links consistent with copper twin axial cables, with lengths up to at least 3m • Define a single-lane 25 Gb/s PHY for operation over links consistent with copper twin axial cables, with lengths up to at least 5m • Define a single-lane 25 Gb/s PHY for operation over MMF consistent with IEEE P802.3bm Clause 95 The IEEE 802.3by standard was approved in June 2016. In addition to the IEEE 802.3by 25 Gb/s Ethernet Task Force an industry consortium (the 25G Ethernet Consortium) was formed in July 2014 by Arista, Broadcom, Google, Mellanox and Microsoft to support the specification of single www.xenanetworks.com 5 WHITE PAPER lane 25 Gigabit/s Ethernet and dual lane 50 Gigabit/s Ethernet technology. The 25G Ethernet Consortium draft specification was completed in September 2015. The following table is based on a table in the 25G Ethernet Consortium specification showing a summary of the two solutions: Port Mode Lane Speed per Lane PCS Lane Bonding Auto- Count (GHz) negotiation A. Single Lane 25G 1 25.78125 N/A Yes B. Two lane 50G 2 25.78125 MLD4 or RS FEC (see Yes end of next section) Table 1: 25G and 50G solutions summary In 2016 IEEE initiated work on defining 50 Gigabit Ethernet by setting up the IEEE 802.3cd task force. The objective of the task force is however to define 50 Gigabit Ethernet over a single lane instead of the dual lane solution as outlined in the 25G Ethernet Consortium specification as a single lane technology is expected to be more cost effective than the dual lane technology. To achieve this it will be necessary to use a Pulse Amplitude Modulation technique PAM-4, where 4 different pulse amplitudes are used to transfer the information. The IEEE 802.3cd task force target is to complete the 50 Gigabit Ethernet single lane standard in the 2018-2019 time frame. PROTOCOL STACK The Ethernet protocol stack is based on the seven-layer OSI model of computer networking: Layer 7 Application layer Layer 6 Presentation layer Layer 5 Session layer Layer 4 Transport layer Layer 3 Network layer Layer 2 Data link layer Layer 1 Physical layer Table 2: The OSI model of computer networking The 25G Ethernet Consortium specification for 25 Gigabit Ethernet (25GE) and 50 Gigabit Ethernet (50GE) addresses Layer 1, which contains the following sublayers: www.xenanetworks.com 6 WHITE PAPER Sublayer Functions • PCS (Physical Coding auto-negotiation •
Load more

Recommended publications
  • On Ttethernet for Integrated Fault-Tolerant Spacecraft Networks
    On TTEthernet for Integrated Fault-Tolerant Spacecraft Networks Andrew Loveless∗ NASA Johnson Space Center, Houston, TX, 77058 There has recently been a push for adopting integrated modular avionics (IMA) princi- ples in designing spacecraft architectures. This consolidation of multiple vehicle functions to shared computing platforms can significantly reduce spacecraft cost, weight, and de- sign complexity. Ethernet technology is attractive for inclusion in more integrated avionic systems due to its high speed, flexibility, and the availability of inexpensive commercial off-the-shelf (COTS) components. Furthermore, Ethernet can be augmented with a variety of quality of service (QoS) enhancements that enable its use for transmitting critical data. TTEthernet introduces a decentralized clock synchronization paradigm enabling the use of time-triggered Ethernet messaging appropriate for hard real-time applications. TTEther- net can also provide two forms of event-driven communication, therefore accommodating the full spectrum of traffic criticality levels required in IMA architectures. This paper explores the application of TTEthernet technology to future IMA spacecraft architectures as part of the Avionics and Software (A&S) project chartered by NASA's Advanced Ex- ploration Systems (AES) program. Nomenclature A&S = Avionics and Software Project AA2 = Ascent Abort 2 AES = Advanced Exploration Systems Program ANTARES = Advanced NASA Technology Architecture for Exploration Studies API = Application Program Interface ARM = Asteroid Redirect Mission
    [Show full text]
  • 40 and 100 Gigabit Ethernet Overview
    Extreme Networks White Paper 40 and 100 Gigabit Ethernet Overview Abstract This paper takes a look at the main forces that are driving Ethernet bandwidth upwards. It looks at the standards and architectural practices adopted by the different segments, how the different speeds of Ethernet are used and how its popularity has resulted in an ecosystem spanning data centers, carrier networks, enterprise networks, and consumers. Make Your Network Mobile © 2011 Extreme Networks, Inc. All rights reserved. Do not reproduce. Extreme Networks White Paper: 40 and 100 Gigabit Ethernet Overview and how its popularity has resulted in a complex ecosys- Overview tem between carrier networks, enterprise networks, and consumers. There are many reasons driving the need for higher bandwidth Ethernet, however, the main reason is our insatiable appetite for content. The definition of content Driving the Need for Speed in itself has evolved over time – where once the majority of traffic on an Ethernet network may have been occa- Ethernet in the Enterprise and Data sional file transfers, emails and the like, today technology Center is allowing us to push and receive richer content such Data center virtualization, which includes storage and as voice, video and high definition multimedia. Simi- server virtualization, is all about the efficient use of larly, mechanisms for delivering content have evolved resources. In the data center this is multifaceted. On over time to reflect this demand. While there were a few the one hand data center managers are trying to bring technologies competing for LAN dominance in the early power, cooling and space utilization under control, while days of networks, Ethernet has become the clear choice.
    [Show full text]
  • ATM Vs Gigabit Ethernet: a Technical Perspective
    White Paper Gigabit Ethernet and ATM A Technology Perspective Bursty, high-bandwidth applications are driving the need for similarly high-bandwidth campus backbone infrastructures. Today, there are two choices for the high-speed campus backbone: ATM or Gigabit Ethernet. For many reasons, business and technical, Gigabit Ethernet is selected as the technology of choice. This paper briefly presents, from a technical perspective, why Gigabit Ethernet is favored for most enterprise LANs. In the past, most campuses use shared-media backbones — such as 16/32 Mbps Token-Ring and 100 Mbps FDDI — that are only slightly higher in speed than the LANs and end stations they interconnect. This has caused severe congestion in the campus backbones when these backbones interconnect a number of access LANs. A high capacity, high performance, and highly resilient backbone is needed-one that can be scaled as end stations grow in number or demand more bandwidth. Also needed is the ability to support differentiated service levels (Quality of Service or QoS), so that high priority, time-sensitive, and mission-critical applications can share the same network infrastructure as those that require only best-effort service. 2 Gigabit Ethernet and ATM: A Technology Perspective White Paper Until recently, Asynchronous Transfer Interface, and Multiprotocol over ATM). Which is a “better” technology is no Mode (ATM) was the only switching This additional complexity is required in longer a subject of heated industry debate technology able to deliver high capacity order to adapt ATM to the connectionless, — Gigabit Ethernet is an appropriate and scalable bandwidth, with the promise frame-based world of the campus LAN.
    [Show full text]
  • 100 Gigabit Ethernet Is Here!
    100 Gigabit Ethernet is Here! Introduction Ethernet technology has come a long way since its humble beginning in 1973 at Xerox PARC. With each subsequent iteration, there has been a lag between time of standardization and large scale adoption. The latest iteration, dubbed 802.3ba by the IEEE Higher Speed Study Group (HSSG), was ratified in June, 2010 and follows this same pattern but with a slight twist. For the first time in Ethernet history a single standard defines two separate speeds; 100 Gigabit Ethernet (100GbE) as well as 40 Gigabit Ethernet (40GbE). Figure 1: Original Ethernet Sketch The technical challenges facing 100GbE have been significant; ranging from developing a whole new generation of optics that can handle 4 lanes of 25Gbps, to simply dealing with normal router and switch functions such as packet inspection, queuing, lookups, filtering and table updates, all in one-tenth the amount of time as with 10GbE. And of course this all has to be done with complete backwards compatibility and meeting all expectations with respect to bit error rate, latency, jitter and the like. As expected 40GbE gained some level of market acceptance first, but some 5 years after ratification the time for 100 Gigabit Ethernet is now! 2 | P a g e This whitepaper will discuss the evolution of 100GbE technology in the service provider and data center markets and provide insights in to how network application acceleration hardware can be leveraged to maximize performance and efficiency in emerging 100GbE network appliances. 100GbE in Service Providers Networks 100GbE is rapidly approaching large scale adoption in the wide area network (WAN), which is largely the purview of service providers.
    [Show full text]
  • Transceiver Product Guide
    PUBLIC_REV2017_N Transceiver Product Guide TRANSCEIVER PRODUCT GUIDE Skylaneoptics.com Transceivers for Datacom and Telecom Applications Skylane Optics is a leading provider of transceivers for optical communication. We offer an extensive portfolio for the enterprise, access, and metropolitan fiber optical market. The offerings provided by Skylane Optics are characterized by high quality and performance. In combination with our strong technical support, we enable our customers to build cost optimized network solutions solving existing and future capacity needs. Solutions Data Center Optimized fiber optic solution for Data Center Application FTTH Broad Product Portfoloio and Technology for FTTH Broadband Networks Wireless Enabling Rapid Expnsion of Mobile Broadband Enterprise - Campus We provides the enterprise network market with the most comprehensive product combinations TRANSCEIVER PRODUCT GUIDE P01 Products Our Engineering and Logistics Center > Inventory, logistics, programming and quality > control based in Fraire, Belgium > IQC [Incoming Quality Control] and OQC > [Outgoing Quality Control] > 100% optimized for handling of transceivers > SD [ANSI/ESD S20.20] compliant > Clean room environment; class 100K > Traceability > High Capacity Our Laboratory > Lab, based in Fraire, Belgium > Technical support > RMA handling > Qualification tests: > - Measure performance over the temperature range to verify compliance with standards > - Compliance with standards (IEEE, IEC, MSA) > - Power consumption > - Eye diagram > - Sensitivity > - Wavelength TRANSCEIVER PRODUCT GUIDE P02 Why Skylane Optics ? Innovations for Early Adopters Quality & Assurance Customization The manufacturing environment is strictly We have cutting-edge test equipment to Due to our high experienced engineers, compliant to most avanced standard, which ensure we supply high quality products. we are enable to modify the hardware and ensure long term reliability. software of the transceivers.
    [Show full text]
  • DECLARATION of CONFORMITY Manufacturer: Fiberworks AS
    DECLARATION OF CONFORMITY Manufacturer: Fiberworks AS Manufacturer's Address: Eikenga 11, 0579 Oslo, Norway declare, under its sole responsibility that the products listed in Appendix I conforms to RoHS Directive 2011/65/EU and EMC Directive 2014/30/EU tested according to the following standards: ROHS EMC IEC 62321-1:2013 EN 55032:2015 IEC 62321-3-1:2013 EN55035:2017 IEC 62321-4:2013 EN 61000-3-2:2014 IEC 62321-5:2013 EN 61000-3-3:2013 IEC 62321-6:2015 IEC 62321-7-1:2015 IEC 62321-8:2017 and thereby can carry the CE and RoHS marking. Ole Evju Product Manager Transceivers Fiberworks AS Fiberworks AS, Eikenga 11, 0579 Oslo, Norway - www.fiberworks.no APPENDIX I SFP SFP SFP BiDi STM-1 (155 Mbps) 4x/2x/1x Fibre Channel Fast Ethernet Bi-Di w/DDM SFP-155M-L15D SFP-4GFC-SD SFP-FE-BX20D-34 SFP-155M-L40D SFP-4GFC-L5D SFP-FE-BX20D-43 SFP-155M-L80D SFP-4GFC-L10D SFP-MR155-BX20D-35 STM-1 (155 Mbps) xWDM SFP-4GFC-ER SFP-MR155-BX20D-53 SFP-155M-L80D-Cxx SFP-4GFC-ZR SFP-MR155-BX40D-35 SFP-155M-L160D-Cxx 4x/2x/1x Fibre Channel xWDM SFP-MR155-BX40D-53 Fast Ethernet SFP-4GFC-30D-Cxx Fast Ethernet Bi-Di w/o DDM SFP-100Base-FX SFP-4GFC-50D-Cxx SFP-FE-BX2D-35 SFP-100Base-LX SFP-4GFC-80D-Cxx SFP-FE-BX2D-53 SFP-100Base-EX SFP-4GFC-80D-Dxxxx SFP-FE-100BX-U-10 SFP-100Base-ZX SFP-FE-100BX-D-10 SFP-100Base-L160D SFP Copper SFP-MR155-BX40-35 SFP-GE-FE-FX SFP-1000Base-TX SFP-MR155-BX40-53 SFP-GE-FE-LX SFP-GE-T Dual Rate 100/1000Mb Fast Ethernet xWDM SFP-100Base-T 1310/1490 SFP-100B-L40D-Cxx SFP-DR-BX20D-34 SFP-100B-L80D-Cxx SFP-DR-BX20D-43 SFP-100B-L160D-Cxx SFP-DR-BX40D-34
    [Show full text]
  • IEEE Std 802.3™-2012 New York, NY 10016-5997 (Revision of USA IEEE Std 802.3-2008)
    IEEE Standard for Ethernet IEEE Computer Society Sponsored by the LAN/MAN Standards Committee IEEE 3 Park Avenue IEEE Std 802.3™-2012 New York, NY 10016-5997 (Revision of USA IEEE Std 802.3-2008) 28 December 2012 IEEE Std 802.3™-2012 (Revision of IEEE Std 802.3-2008) IEEE Standard for Ethernet Sponsor LAN/MAN Standards Committee of the IEEE Computer Society Approved 30 August 2012 IEEE-SA Standard Board Abstract: Ethernet local area network operation is specified for selected speeds of operation from 1 Mb/s to 100 Gb/s using a common media access control (MAC) specification and management information base (MIB). The Carrier Sense Multiple Access with Collision Detection (CSMA/CD) MAC protocol specifies shared medium (half duplex) operation, as well as full duplex operation. Speed specific Media Independent Interfaces (MIIs) allow use of selected Physical Layer devices (PHY) for operation over coaxial, twisted-pair or fiber optic cables. System considerations for multisegment shared access networks describe the use of Repeaters that are defined for operational speeds up to 1000 Mb/s. Local Area Network (LAN) operation is supported at all speeds. Other specified capabilities include various PHY types for access networks, PHYs suitable for metropolitan area network applications, and the provision of power over selected twisted-pair PHY types. Keywords: 10BASE; 100BASE; 1000BASE; 10GBASE; 40GBASE; 100GBASE; 10 Gigabit Ethernet; 40 Gigabit Ethernet; 100 Gigabit Ethernet; attachment unit interface; AUI; Auto Negotiation; Backplane Ethernet; data processing; DTE Power via the MDI; EPON; Ethernet; Ethernet in the First Mile; Ethernet passive optical network; Fast Ethernet; Gigabit Ethernet; GMII; information exchange; IEEE 802.3; local area network; management; medium dependent interface; media independent interface; MDI; MIB; MII; PHY; physical coding sublayer; Physical Layer; physical medium attachment; PMA; Power over Ethernet; repeater; type field; VLAN TAG; XGMII The Institute of Electrical and Electronics Engineers, Inc.
    [Show full text]
  • The Future Is 40 Gigabit Ethernet White Paper Cisco Public
    The Future Is 40 Gigabit Ethernet White Paper Cisco Public The Future Is 40 Gigabit Ethernet © 2016 Cisco and/or its affiliates. All rights reserved. The Future Is 40 Gigabit Ethernet White Paper Cisco Public Executive Summary The business case for 40 Gigabit Ethernet is becoming inescapably compelling. While 10 Gigabit Ethernet is still making its way into the data centers, CIOs and IT managers must now consider how they are going to handle what’s coming next: high-bandwidth applications such as server virtualization and cloud computing; fabric consolidation within the data center; and a greater demand for high-performance computing among end users (see Figure 1). The need for faster data transfer rates is relentless and carries significant implications with regard to network productivity as well as operating expenditure (OpEx) costs. Figure 1. Current Trends Driving the Demand for This report addresses the impending move to 40 Higher-Speed Ethernet Gigabit Ethernet, how it may change the network architecture, and what IT managers can do now to Market Drivers for More Bandwidth prepare to migrate to the new standard. Consumer & Broadband Access Introduction: The Business Case for Content 40 Gigabit Ethernet Providers Since February 1980, when the first IEEE 802 Server Virtualization standards committee convened, speeds in Ethernet Video on delivery to all layers have made increasingly greater Demand leaps over increasingly shorter intervals. In 2016, Blade Server Higher eight years after the adoption of 10 Gigabit Ethernet, Speed Service the IEEE has adopted 802.3ba, paving the way for Providers & Ethernet IXCs 40 Gigabit Ethernet and 100 Gigabit Ethernet.
    [Show full text]
  • 40 and 100 Gigabit Ethernet: an Imminent Reality
    WHITE PAPER 40 and 100 Gigabit Ethernet: An Imminent Reality 40 and 100 Gigabit Ethernet: An Imminent Reality Many of today’s data centers are running 10 Gigabit Ethernet (GbE) over both optical fiber and balanced twisted-pair copper cabling in their backbone infrastructure where large numbers of gigabit links aggregate at core devices. As more edge devices; like servers and storage equipment, continue to move to 10 GbE, the next natural progression is for the network core to require even faster connections within the data center. Fortunately, there is a solution that is now an imminent reality. Standards have been in development since 2008, and the Institute of Electrical and Electronics Engineers (IEEE) will soon release the 802.3ba standard that will support data rates for 40 and 100 GbE over optical fiber cabling. Both cable and connectivity solutions capable of supporting these speeds already exist, and vendors are in the process of developing active equipment. Now is the time to migrate data center cabling infrastructures to support this imminent technology. 40 and 100 Gigabit Ethernet: An Imminent Reality Key Market Drivers 100 90 From storage and IP traffic growth to the advancement 35% CAGR in Storage Capacity of technology across many market sectors, the drivers 80 that moved data transmission speeds from 1 GbE to 68 10 GbE over the past decade are now expanding as 60 forecasted, creating the need for 40 and 100 GbE. 49 Petabytes 40 37 10 GbE Growth 28 20 20 While the global Ethernet switch market experienced overall decline in 2009, the migration from 1 to 10 0 GbE continued in data centers across the world.
    [Show full text]
  • Towards 100 Gbps Ethernet: Development of Ethernet / Physical Layer Aspects
    SEMINAR ON TOPICS IN COMMUNICATIONS ENGINEERING 1 Towards 100 Gbps Ethernet: Development of Ethernet / Physical Layer Aspects Ömer Bulakci Abstract — Physical layer features of Ethernet from the first released clauses and ongoing architecture researches for 100 realization towards the 100 Gb Ethernet (100 GbE) development GbE are elaborated. have been considered. Comparisons of these features are made according to the standardized data rates. Feasible physical layer TABLE I options are then discussed for high data rates. Milestones of 802.3 IEEE Standard I. INTRODUCTION Clause Date of Bit Physical THERNET is the most widely deployed Local Area Name Release Rate Medium Network (LAN) protocol and has been extended to E 802.3a Single Metropolitan Area Networks (MAN) and Wide Area (Thin Ethernet) 1985 10 Mbps Thin Coaxial Networks (WAN) [1]. The major advantages that characterize (Cheapernet) Cable Ethernet can be stated as its cost efficiency, traditional tenfold bit rate increase (from 10 Mbps to 100 Gbps), simplicity, high 802.3i 1990 10 Mbps TP Copper transmission reliability and worldwide interoperability 802.3j 1993 10 Mbps Two MMFs between vendors [2]. TP Copper The first experimental Ethernet was developed during the 802.3u 1995 100 Mbps Two Fibers early 1970s by XEROX Corporation in a coaxial cable (Fast Ethernet) (MMF,SMF) network with a data rate about 3 Mbps [3]. The initial 802.3z 1998 1 Gbps MMF, SMF standardization process of Ethernet was started in 1979 by (Gigabit Ethernet) Digital Equipment Corporation (DEC), Intel and Xerox. In 802.3ab 1999 1 Gbps TP Copper 1980, DIX Standard known as the “Thick Ethernet” was 802.3ae 2002 10 Gbps MMF,SMF released.
    [Show full text]
  • 25G Ethernet CFI
    25G Ethernet CFI Final Draft Brad Booth, Microsoft Photo courtesy of Hugh Barrass Objectives • To gauge the interest in starting a study group to investigate a 25 Gigabit Ethernet project • Don’t need to: • Fully explore the problem • Debate strengths and weaknesses of solutions • Choose a solution • Create a PAR or 5 Criteria • Create a standard • Anyone in the room may vote/speak 2 25G Ethernet CFI Agenda • Overview • MAC-PHY Mismatch • Potential Use Cases • Why Now? • Straw Polls 3 25G Ethernet CFI Agenda • Overview • MAC-PHY Mismatch • Potential Use Cases • Why Now? • Straw Polls 4 25G Ethernet CFI 25G Ethernet Overview • Provide a 25G media access control (MAC) that matches the single-lane 25G physical layer (PHY) technology • In web-scale data centers, 25G Ethernet could provide an efficient server to top-of-rack (TOR) speed increase • Predominantly direct-attach copper (DAC) cable • The speed of the PCIe host bus is not moving as fast as networking connectivity speeds 5 25G Ethernet CFI Existing 10G Topology TOR Switch ASIC • Today’s volume topology 48-port 10G 4-port 40G for web-scale data centers • 48 servers/TOR Server • 3:1 oversubscription Server • Uses low-cost, thin 4-wire SFP+ DAC cable Server 6 25G Ethernet CFI Existing 4x10G Topology TOR Switch ASIC • Commonly used topology in web-scale data centers 48-port 10G 4-port 40G • Permits non-blocking 10G mesh • 40G ports used as 4x10G out cable out - Server with QSFP+ to SFP+ break- Server Server out cable Break Server • Same server network interface card (NIC) as 10G Server 7 25G Ethernet CFI 40G Topology TOR Switch ASIC • High-performance, low- 32-port 40G volume topology • Uses bulkier 16-wire QSFP+ DAC cable Server • Max.
    [Show full text]
  • IEEE Standard for Ethernet
    IEEE Standard for Ethernet Amendment 11: Physical Layer and Management Parameters for Serial 25 Gb/s Ethernet Operation Over Single-Mode Fiber IEEE Computer Society Sponsored by the LAN/MAN Standards Committee IEEE IEEE Std 802.3cc™-2017 3 Park Avenue (Amendment to New York, NY 10016-5997 IEEE Std 802.3™-2015 USA as amended by IEEE Std 802.3bw™-2015, IEEE Std 802.3by™-2016, IEEE Std 802.3bq™-2016, IEEE Std 802.3bp™-2016, IEEE Std 802.3br™-2016, IEEE Std 802.3bn™-2016, IEEE Std 802.3bz™-2016, IEEE Std 802.3bu™-2016, IEEE Std 802.3bv™-2017, IEEE Std 802.3-2015/Cor 1-2017, and IEEE Std 802.3bs™-2017) IEEE Std 802.3cc™-2017 (Amendment to IEEE Std 802.3™-2015 as amended by IEEE Std 802.3bw™-2015, IEEE Std 802.3by™-2016, IEEE Std 802.3bq™-2016, IEEE Std 802.3bp™-2016, IEEE Std 802.3br™-2016, IEEE Std 802.3bn™-2016, IEEE Std 802.3bz™-2016, IEEE Std 802.3bu™-2016, IEEE Std 802.3bv™-2017, IEEE Std 802.3-2015/Cor 1-2017, and IEEE Std 802.3bs™-2017) IEEE Standard for Ethernet Amendment 11: Physical Layer and Management Parameters for Serial 25 Gb/s Ethernet Operation Over Single-Mode Fiber LAN/MAN Standards Committee of the IEEE Computer Society Approved 6 December 2017 IEEE-SA Standards Board Abstract: This amendment to IEEE Std 802.3-2015 adds Physical Layer (PHY) specifications and management parameters for 25 Gb/s operation over single-mode fiber at reaches of at least 10 km (25GBASE-LR) and 40 km (25GBASE-ER).
    [Show full text]