A Big Ethernet World: 10M to 400G

We are living in a connected world and there are over a billion Ethernet devices around the world today. It is very interesting how Ethernet has evolved from a simple standard supporting 10 Mbps to multitude of speed modes and ubiquitous applications. From being a standard that traditionally evolved with 10x speed jumps (10M, 100M, 1G, 10G, 100G), it is now growing rapidly at non-10x speed modes (2.5G, 25G, 40G, 50G, 200G, and latest 400G) and covering diverse application areas to satisfy consumer needs.

Join us to go around the big world of Ethernet in the upcoming blogs on the intricacies of the standard IEEE 802.3 – 2012 and draft committees working on 25/50/400G, its interpretations, common gotchas; defining verification strategies, writing test cases, UNH IOL Compliance suite, and so on and so forth. Let’s start the journey with this blog on early days of Ethernet. Bon Voyage.

Ethernet development started around 1974 and was commercially introduced as a DIX standard (acronym for DEC/Intel/Xerox) in 1980. It was made a standard as IEEE 802.3 in 1983 and since then, Ethernet has evolved to meet new bandwidth and market requirements.

It all started with a shared medium to send and receive data traffic over a half-duplex channel (only transmit or receive at a time). Installation was costly, reliability of data transfer was less and troubleshooting was tough. CSMA/CD (Carrier Sense Multiple Access with Collision Detection) was used as the mechanism to share the medium. Collisions detection and retransmission of data was enabled with the help of layers above Data Link layer. Excessive collisions were costly due to reduced throughput and system reset requirement in case of congestion on the medium. Modern Ethernet is full-duplex (can transmit and receive simultaneously) and the medium is no longer shared. Bandwidth gets fully utilized to obtain maximum data throughputs. The medium used to transfer data can be either coaxial, twisted pair or fiber optic. Ethernet standards continue to embrace new media, higher transmission speeds and additions to frame content with advent of time.

Depending on the area that Ethernet covers, it has been divided on 3 categories: Local Area Network (LAN), Metropolitan Area Network (MAN) and Wide Area Network (WAN). The computer networking technology evolved on the 7-layer ISO OSI (Open System Interconnection) reference model, which helped in standardizing communication between layers to provide an error-free path across a network. Ethernet standard provides services up to Data Link Layer, i.e. it covers the Physical layer (layer 1) and Data Link Layer (Layer 2), in the 7-layer OSI model. We will discuss more about the OSI reference model in detail in upcoming blog.

Stay tuned for upcoming blogs on Ethernet covering following topics in length and breadth (but will not limit ourselves to explore more).

• Medium Access Control sublayer (MAC) for 10/100M, 1G: Clauses 4, 35
• Medium Access Control sublayer (MAC) for 10/40/100G: Clauses 46, 81
• Medium Access Control sublayer (MAC) for 25/50/400G: Clauses 106, 117
• Physical Coding sublayer (PCS) for 1G Base-X: Clause 36
• Auto-negotiation: Clause 37
• Management Data Input Output (MDIO): Clauses 22, 45
• Physical Coding sublayer (PCS) for 10G Base-X: Clause 48
• Physical Coding sublayer (PCS) for 10G Base-R: Clause 49
• Ethernet over Backplane
• Auto-adaptation: Clause 72
• Auto-negotiation: Clause 73
• Forward Error Correction (FEC) for Base-R: Clause 74
• Reed Solomon FEC (RS-FEC): Clause 108
• Physical Coding sublayer (PCS) for 40/100G Base-R: Clause 82
• Energy Efficient Ethernet (EEE): Clause 78
• Consortium 25/50G
• IEEE 25/50G
• Physical Coding Sub-layer (PCS) for 400G Base-R: Clause 119
• MAC interfaces: MII/GMII/RMII/RGMII/XGMII/XLGMII/CGMII/CDMII
• Upper layer packets: Layer 2, 2.5, 3, 4 and 7: Include, VLAN, GRE, IPv4, IPv6, TCP, UDP, Tagged frames, etc.
• IEEE 1588 Precision Time Protocol (PTP)
• SGMII
• QSGMII

Synopsys: Your Verification Partner for Next-Gen Networking and Communication SoCs

Synopsys VC VIP for Ethernet is implemented in native SystemVerilog/UVM architecture,  and provides built-in coverage, protocol checks, and Verdi protocol-aware debug. Synopsys VC VIP is capable of switching speed configurations dynamically at run time, and includes an extensive and customizable set of frame generation and error injection capabilities. In addition, source code UNH-IOL test suites are also available for key Ethernet features and clauses, allowing teams to quickly jump start their own custom testing and achieve accelerated verification closure.

Synopsys provides a comprehensive Ethernet solution for all speeds, including 25G, 40G, 50G, 100G and recently announced the availability of the industry’s first verification IP (VIP) and source code test suite to support the proposed IEEE P802.3bs/D1.0 Ethernet 400G standard.

Authored by Jaspreet Singh Gambhir