Link Layer Evaluation

The simulation of the link layer can establish an out-of-line crosstalk test model by simulating “6-pack 1” or “12-pack 1”, and sending and receiving 10Gbps simulated data traffic for metering at both ends of the core cable (victim cable), and select some test items in RFC 2544, through a certain period of continuous testing, get the bit error rate, frame loss rate, and packet loss rate before and after the out-of-line crosstalk interference to complete the certification test.

RFC 2544 testing can last for a long time, up to several hours depending on the setup. The instrument can estimate the test time according to the selected test item. According to the standard, the bit error rate of 10GBase-T 10 Gigabit copper cable is less than or equal to 10-12 in all transmission distances and grades. From this calculation, the test time of the meter should not be less than 103 seconds (ie the minimum test time is 17 minutes). If the standard requirements are met, the cable can be considered to meet the standard, and if the test fails, fault analysis and troubleshooting are required.

Troubleshooting

In practical projects, since the wiring on one side of the working area has been dispersed, the alien crosstalk can be ignored (other electromagnetic interference from the outside is considered separately). The alien crosstalk mainly comes from horizontal wiring arranged in bundles, dense wiring in the cabinet, and wiring at the terminals where the racks are close to each other.

Horizontal cabling alien crosstalk is related to the number and distance of adjacent cables and connectors. The alien crosstalk link segment indicators: PS ANEXT and PS AELFEXT, are measured based on the “6-in-1” model. Alien crosstalk coupling is a function of the distance between wiring elements, and measures such as moving or loosening cables, dispersing equipment or jumpers, and using low-density connectors can significantly reduce alien crosstalk. In many practical cabling topologies, cables are only bundled over a relatively short distance or placed in cable trays, and the cables of star topology cabling are randomly distributed to each work area by the central equipment. The distance between the link segments close to each other is reduced, which is beneficial to reducing the alien crosstalk.

When the alien crosstalk index does not meet the requirements, it can be improved by the following methods (IEEE 802.3an Annex 55B):

(1) When conditions permit, the position of the patch panel between the scattered equipment, the adjacent position can be used for other applications.

(2) Connect accessory equipment to horizontal wiring through internal connections instead of cross-connects to reduce the number of Colacated connectors. Connectors and jumpers in the work area are not Co-located connectors.

(3) By dispersing jumpers and unbundling horizontal wiring, the alien crosstalk coupling of the first 5m~20m horizontal wiring is reduced. The obvious part of the alien crosstalk coupling generally occurs within 20m of the wiring.

(4) Disperse the jumpers of the equipment to fully alleviate the coupling of alien crosstalk.

(5) Replace the connector with a higher type or grade.

This paper mainly discusses how to test the alien crosstalk of unshielded cables of Category 6A and above, and analyzes it from the link-layer simulation. In a word, the cabling engineering of the data center is relatively complicated, and various factors need to be considered comprehensively to ensure that the 10GBASE-T 10 Gigabit copper cable fully reflects its superior characteristics in terms of transmission performance.

LEAVE A REPLY

Please enter your comment!
Please enter your name here