tag:blogger.com,1999:blog-432335505061070582011-05-03T01:54:18.582-07:00From Bits to PacketsJust share personal understanding on network technology from bits to packetsl3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.comBlogger18125tag:blogger.com,1999:blog-43233550506107058.post-79545116574507448272011-03-01T11:34:00.000-08:002011-03-01T11:34:57.951-08:002011-03-01T11:34:57.951-08:00Cisco's Innovation in Data Center ??I love the diagram in the following link, <br />
<br />
<a href="http://blogs.cisco.com/news/ciscos-unified-fabric-in-the-data-center/">Cisco’s Unified Fabric in the Data Center</a><br />
<br />
which states the innovative technology<br />
<br />
Cisco did in Data Center domain.<div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-7954511657450744827?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-67645102184162731982010-12-24T00:33:00.000-08:002010-12-24T00:33:21.720-08:002010-12-24T00:33:21.720-08:00802.1p vs 802.1Q802.1p defines the class of service (Cos) in Ethernet MAC layer, and uses 802.1Q's 3-bit field to indicate 8 different traffic priority. The higher the value, the higher priority the packet has. When the packet is tagged with VLAN 0, it means it is 802.1p packets only.<br />
802.1Q defines the virtual LAN tagging method in Ethernet MAC layer and is used to divide one physical network into multiple virtual LANs. It adds 4 bytes into original Ethernet frame. 2 bytes are used to indicate 802.1Q packets, which are 0x8100. The left 2 bytes consist of 3 fields: 3-bit Priority Code Point (PCP), which is defined by 802.1p; 1-bit Canonical Format Indicator (CFI), which indicate whether it is Ethernet or Token Ring; 12-bit VLAN identifier, which is used to indicate which VLAN the packet belongs to. VLAN 0 is used by 802.1p to deliver Cos priority only, while 4096 is reserved. The available VLAN range is 1 ~ 4095 .<div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-6764510218416273198?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-38422090404701990422010-12-21T21:22:00.000-08:002010-12-21T21:30:23.873-08:002010-12-21T21:30:23.873-08:00Gb/s, GB/s and GT/sIn data communication or computer industry, Gb/s, GB/s or GT/s is used to describe the data transferring rate or bandwidth between endpoints. Gb/s is gigabits per second, GB/s is gigabytes per second, while GT/s is gigatransfers per second. The conversion between Gb/s and GB/s is simple: dividing Gb/s by 8 will get GB/s. <br />
To convert GT/s to Gb/s, the encoding of the data has to be looked at. For example, PCIe 2.0, it mentions the interconnect bit rate is 5 GT/s. Considering that encoding scheme in PCIe is 8b/1b, the effective bit rate is 5 Gb/s * (8/10) = 4 Gb/s. Basically, GT/s is raw data rate, while Gb/s is effective data rate.<div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-3842209040470199042?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-67227849313323865672010-12-21T19:50:00.000-08:002010-12-21T20:14:55.919-08:002010-12-21T20:14:55.919-08:00Terminal ServerTerminal server (console server) is one device which connected to multiple devices (switch, router, server, or printer)using RS-232 connector on one end, and connected to Ethernet via RJ-45 connector on other end. Thus, system administrators are able to access and configure those devices remotely via IP network. Terminal server is one converter between serial and Ethernet.<div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-6722784931332386567?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-31995481525100186492010-12-21T15:49:00.000-08:002010-12-21T15:58:47.907-08:002010-12-21T15:58:47.907-08:0019 Inch Rack and Rack Unit (RU)In telecommunication, computer network or audio/video industry, 19-inch rack chassis cabinet is widely used.19-inch means the width of the equipment front panel, and not the distance between holes.19 inch is about 48.26cm or less than half one meter. As to how tall the chassis is , it is measured in Rack Unit (RU).Generally, one chassis could hold multiple RUs. Here, 1 RU equals to1.75 inch, which is about 44.45 mm. 1.75 inch is the not height of front panel. The fact is the height of the front panel is 1/32 inch less --- about (44.45 - 0.79) = 43.66 mm. Such design will give some spaces between RUs in the rack.<br />
<br />
When buying chassis based switch/routers or servers, good to buy those standard chassis. When looking into<br />
the data sheets or production information next time, you know what those terminology means. <br />
<br />
23 inch or 24 in rack also exist in industry, but not as popular as 19 inch rack chassis.<div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-3199548152510018649?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-10500989931552777982010-12-20T20:08:00.000-08:002010-12-21T19:33:00.439-08:002010-12-21T19:33:00.439-08:00Technology Websites<ol><li><a href="http://www.eetimes.com/">EE Times</a></li>
<li><a href="http://www.nullmodem.com/">NullModerm </a></li>
<li><a href="http://www.lightreading.com/">Light Reading</a></li>
<li><a href="http://www.javvin.com/">Javvin </a></li>
<li><a href="http://www.techcrunch.com/">Tech Crunch</a></li>
<li><a href="http://www.design-reuse.com/">Design & Resuse</a></li>
</ol><div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-1050098993155277798?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-74290673560965366472010-12-20T10:35:00.000-08:002010-12-21T16:07:42.443-08:002010-12-21T16:07:42.443-08:00BPS (Bits per Second) verus PPS ( Packet per Second)BPS is also written as b/s. Both BPS and PPS are used to describe the network bandwidth usage. The relationship between BPS and PPS depends on the packet size. However, the defacto packet size in industry is 64 bytes. To convert PPS to BPS using the formula: BPS = PPS * (64 * 8 ) bits<br />
<br />
bps has Kilo bits per second (Kbps), Mega bits per second (Mbps), Giga bits per second (Gbps) and Tera bits per second (Tbps); While pps has Kilo packets per second (Kpps), Maga packets per second (Mpps), Billion packets per second (Bpps) . <br />
<br />
The reason why we need pps is that switch/router's packet processing capability depends on packet size. That is, switch/router can't do wire speed for all packets size, but should be able to process most packet sizes at line rate.<div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-7429067356096536647?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-38705703942575880092010-12-19T20:18:00.000-08:002010-12-21T21:54:56.908-08:002010-12-21T21:54:56.908-08:00Ethernet Tutorial 2<div style="text-align: center;"><a href="http://l3l2.blogspot.com/2010/12/ethernet-tutorial-1.html">PRE <<<</a></div><br />
<ol><li> Ethernet Frames Format </li>
<table border="0" cellpadding="5" cellspacing="0"><tbody> </tbody><caption>Ethernet Frames</caption> <tbody>
<tr> <th>Field</th> <th>DIX Frame </th> <th>IEEE 802.3 Frame </th> </tr>
<tr> <th>Preamble </th> <td> 8 bytes</td> <td> 7 bytes</td></tr>
<tr> <th>Start Frame Delimiter </th> <td> </td> <td> 1 bytes</td></tr>
<tr> <th>Destination Address </th> <td> 6 bytes</td> <td> 6 bytes</td></tr>
<tr> <th>Source Address </th> <td> 6 bytes</td> <td> 6 bytes</td></tr>
<tr> <th>Length </th> <td> </td> <td> 2 bytes</td></tr>
<tr> <th>Type </th> <td> 2 bytes</td> <td> </td></tr>
<tr> <th>LLC </th> <td><br />
</td> <td> Yes</td></tr>
<tr> <th>Data</th> <td> 1500 bytes</td> <td> Yes</td></tr>
<tr><th>Pad</th><td><br />
</td> <td> Yes</td></tr>
<tr><th>CRC/FCS</th><td> 4 bytes</td> <td> 4 bytes</td></tr>
<tr><th>Total</th><td> 1518 bytes</td> <td> 1518bytes</td></tr>
</tbody></table>Note: 100/1000Mbs Ethernet Systems do not need preamble or start frame delimiter fields. <a name='more'></a>
<li> MAC address </li>
Each NIC has a unique, 48-bit MAC address; The first 24 bits identify the manufacture while the 2nd half is assigned by the manufacture.
<li> Ethernet Access </li>
Before Gigabit Ethernet, CSMA/CD is designed to allow fair access on shared media: Listen to avoid collision before transmitting. If collision occurs, retransmission is necessary. In 1997 IEEE 802.3x introduces full-duplex Ethernet, MAC control protocol and PAUSE.
<li> Ethernet Equipment</li>
<ul><li> Hubs/Repeaters Hubs/Repeaters are the central devices in a star topology and enhance the incoming signals to avoid degradation before transmitting them out. Hubs/Repeaters also support SNMP.</li>
<li>Bridges Bridges transfer MAC layer packets from one network to another; and help prevent collisions by creating separate collision domain. To control the traffic, bridges need to build MAC tables and prevent loops using Spanning Tree Protocol</li>
<li>Switches Switches link multiple networks compared to Bridges. Two categories of switches are commonly in use today: Cut through and Store-and-forward. Cut through provide lower latency while store-and-forward provides more intelligent processing of MAC layer packets.</li>
<li>Routers Routers filter out unnecessary traffic by looking up IP address instead of MAC address in bridges/switches. Routers works like firewall to some extent and provide security functionality.</li>
</ul></ol><br />
<div style="text-align: center;">>>>NEXT</div><div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-3870570394257588009?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-65154879998957945682010-12-19T09:40:00.000-08:002010-12-22T21:04:08.018-08:002010-12-22T21:04:08.018-08:00Introduction to Cisco's Nexus SwitchesWhile Juniper pieces their MX-Series and EX-Series together to provide a data center solution, Arista Networks designed their EOS switches from scratch purposely for data center and cloud computing, Force10 networks and Brocade Networks are warming up on data center switch markets, as the leader in Ethernet switch market, Cisco already had a end-to-end and market-proven data center solution using their specially engineered Nexus series switches.<br />
There are numerous innovations and new technologies in Cisco Nexus series switch, such as DCB, FCOE, Nexus OS, FabricPath, FEX and Nexus 1000V. One article is necessary to analyze those technologies. Here<br />
is a list of Nexus series switches. You could see its portfolio is the richest in industry. <br />
<br />
<a name='more'></a><ol><li> Cisco Nexus 1000V (Inside Server)</li>
<ul><li>VM-aware software switch</li>
<li>Provides Cisco VN-Link VM network services</li>
<li>Work seamlessly with VMware vSphere <iframe align="left" frameborder="0" marginheight="0" marginwidth="0" scrolling="no" src="http://rcm.amazon.com/e/cm?t=fromb-20&o=1&p=8&l=bpl&asins=1587058928&fc1=000000&IS2=1&lt1=_blank&m=amazon&lc1=0000FF&bc1=000000&bg1=FFFFFF&f=ifr" style="height: 245px; padding-right: 10px; padding-top: 5px; width: 131px;"></iframe></li>
</ul><li>Cisco Nexus 2000 Series Fabric Extenders (Access Layer)</li>
<ul><li>1 RU</li>
<li> 48 1000Base-T ports , 4 SFP+ 10GbE uplinks or </li>
<li>24 100/1000Base-T ports, 2 10GbE SFP+ uplinks or</li>
<li>48 100/1000Base-T ports, 4 10GbE SFP+ uplinks or</li>
<li>32 1/10Gb SFP+ Ethernet/FCoE ports, 8 10Gb Ethernet/FCoE uplinks</li>
<li>1+1 Fan Trays</li>
<li>1+1 Power Supplies</li>
<li>vPCs, FCOE</li>
<li>Parent switch could be either Nexus 5000 or Nexus 7000 </li>
</ul><li>Cisco Nexus 5000 Series ( Access Layer) </li>
<ul><li>1 or 2 RU<iframe align="left" frameborder="0" marginheight="0" marginwidth="0" scrolling="no" src="http://rcm.amazon.com/e/cm?t=fromb-20&o=1&p=8&l=bpl&asins=B003E3VTAK&fc1=000000&IS2=1&lt1=_blank&m=amazon&lc1=0000FF&bc1=000000&bg1=FFFFFF&f=ifr" style="height: 245px; padding-right: 10px; padding-top: 5px; width: 131px;"></iframe></li>
<li>32 1/10Gbps SFP+ Etherne /FCoE fixed ports, with expansion </li>
<ul><li>16 10Gbps SFP+; Ethernet/FCoE; </li>
<li> 8 10Gbps SFP+/FCoE and; 8 1/2/4/8Gbps Fibre Channel ports</li>
</ul><li>20 10Gbps SFP+ Ethernet/FCoE fixed ports,with expansion</li>
<ul><li>8 1/2/4Gbps Fibre Channel ports </li>
<li>6 1/2/4/8Gbps Fibre Channel ports</li>
<li> 410Gbps DCB/FCoE Ethernet and 4 1/2/4/Gbps Fibre Channel ports</li>
<li> 610Gbps DCB/FCoE Ethernet </li>
</ul><li>40 10Gbps SFP+ DCB/FCoE Ethernet fixed ports,with 2 expansions</li>
<ul><li>8 1/2/4Gbps Fibre Channel ports </li>
<li>6 1/2/4/8Gbps Fibre Channel ports</li>
<li> 410Gbps DCB/FCoE Ethernet and 4 1/2/4/Gbps Fibre Channel ports</li>
<li> 610Gbps DCB/FCoE Ethernet </li>
</ul><li>1+1 Fan Trays</li>
<li>1+1 Power Supplies</li>
<li>1 or 2 expansion modules </li>
<li>vPCs, DCB, FCoE and Fibre Channel</li>
<li>Parent switch could be Nexus 7000 </li>
</ul><li>Cisco Nexus 7000 Series (Core Layer)</li>
<ul><li> 10 or 18 slot chassis </li>
<li> 8 or 16 I/O modules</li>
<li>32 10GbE I/O module </li>
<li>48 GbE I/O modules</li>
<li>32 1/10GbE I/O Module or</li>
<li>8 10GbE I/O modules</li>
<li>5 Fabric Modules </li>
<li>1+1 Supervisor Module </li>
<li>1+1 System Fan Trays</li>
<li>1+1 Power Supplies</li>
<li>vPCs, VDC, </li>
</ul></ol><div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-6515487999895794568?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-43819865637820810502010-12-18T22:32:00.000-08:002010-12-20T08:21:28.733-08:002010-12-20T08:21:28.733-08:00Introduction to Arista EOS SwitchesArista Networks create a lot of buzz those days. Based on its platform EOS -- another network operating system after Cisco's IOS, Nexus OS and Juniper's JUNOS, Arista Networks delivers a portfolio of Gigabit and 10 Gigabit Ethernet switches targeting Data Center market. <br />
<br />
From its websites, we could know EOS is built on top of Linux and written using C++ and Python. L2/L3 network protocols, CLI, SNMP, and ASIC drivers are running in user space. ProcMgr is the process to do stateful restart and recovery based on System Database (SysDB), which works as IPC mechanism between different user agents. The innovation of EOS is its vEOS, VM Tracer and MLAG.<br />
<br />
In big picture, Arista Networks shares similar view as Cisco's vision on data center network: Access, Aggregation and Core. This may came from the fact that Arista's CEO and founders ever worked in Cisco Systems. The performance looks promising from disclosed data sheet. The per-port pricing is attractive too. However, compared to Cisco's portfolio, the choice Arista Networks provided is limited. The feature set is not that rich. What's more, the solution between leaf switch and server or VM is software-based. Not sure whether it would affect the performance when system scaled bigger.<br />
<a name='more'></a>The following lists switches they provided today.<br />
<ol><li>Arista 7048 (Leaf) </li>
<ul><li>1 RU</li>
<li> 48 1000Base-T ports </li>
<li>4 1/10GbE SFP+ uplink ports</li>
<li>2 redundant hot-swappable power supply</li>
<li>4 redundant hot-swappable fans</li>
<li>768 MB deep buffer</li>
<li>Citrix NetScaler VPX integration </li>
<li>L4-L7 Load Balancing & Security</li>
<li>Arista EOS</li>
</ul><li> Arista 7010 Series(Spine)</li>
<ul><li>1 RU</li>
<li>24 or 48 10Gbs SFP/SFP+ ports </li>
<li>2 redundant hot-swappable power supply</li>
<li>5 redundant hot-swappable fans</li>
<li>2 10/100/1000Mb Mgmt ports</li>
<li>1 RS-232 serial ports</li>
<li>Arista EOS</li>
</ul><li> Arista 7508(Core)</li>
<ul><li>11 RU Chassis with mid-plane design</li>
<li>48 * 8 = 38410Gbs SFP/SFP+ ports </li>
<li>4 redundant hot-swappable power supply</li>
<li>6 fabric modules </li>
<li>2 supervisor modules </li>
<li>6 fan modules</li>
<li>Arista EOS</li>
</ul></ol><div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-4381986563782081050?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-84543240842870075052010-12-18T19:49:00.000-08:002010-12-20T20:49:19.785-08:002010-12-20T20:49:19.785-08:00Ethernet Tutorial 1One Ethernet tutorial I like very much is from Fujisu since it includes all the basics of Ethernet technologies. The following is my reading note, and you could read the original article by clicking <a href="http://www.fujitsu.com/downloads/TEL/fnc/pdfservices/ethernet-prerequisite.pdf">here</a>. You could also<br />
check this<a href="http://www.dcs.gla.ac.uk/%7Eross/Ethernet/index.htm"> link </a>from England.<br />
<ol><li>Ethernet History</li>
Robert Metcalfe from Xerox described his invention Ethernet in 1973. He modeled Ethernet after the Aloha network developed at University of Hawaii in the 1960s and the initial Ethernet systems ran at 2.94Mb/s. Ethernet patent was awarded to Metcalfe and his co-workers, but wholly-owned by Xerox.
<li>Ethernet Standards</li>
In 1980, DEC, Intel and Xerox (DIX) issued a DIX Ethernet standard for 10Mb/s Ethernet systems. In 1985, IEEE published the open Ethernet standard IEEE 802.3. Why 802? It is because work started in February 1980. For all the later Ethernet standard, people could refer Wikipedia link<a href="http://en.wikipedia.org/wiki/IEEE_802.3"> 802.3 .</a>
<li> Ethernet Technologies</li>
</ol><ul><li>Fast Ethernet </li>
</ul> In 1995, IEEE 802.3u defines 100Mb/s Ethernet over wire or fiber-optic cable. Speed could be auto-negotiated between equipments to choose 10Mb/s or 100Mb/s.<br />
<a name='more'></a><br />
<ul><li>Gigabit Ethernet </li>
</ul> Gigabit Ethernet uses same IEEE 802.3 frame format, full duplex and flow control methods. In addition, it supports half-duplex via CSMA/CD and SNMP tools. What's more, it introduces jumbo frames (64 ~ 9216 bytes) to reduce frame rates compared to standard frames (64 ~ 1518 bytes).<br />
Gigabit Ethernet could be transmit over CAT 5 cable or optical fibre such as 1000Base-CX(Copper), 1000 Base-SX(850nm fibre), and 1000Base-LX(1310nm fibre).<br />
<ul><li>10 Gigabit Ethernet </li>
</ul> 10 Gigabit Ethernet is transmitted over multi mode and single mode optical fibre in full-duplex mode, which eliminates the need for CSMA/CD. 10 Gigabit Ethernet defines 2 physical layers: LAN PHY and WAN PHY.<br />
2 versions of LAN PHY are standardized: 10GBase-R and 10GBase-X. 10GBase-R uses a 64B/66B encoding scheme to raise the line rate to 10.313Gb/s; While 10GBase-X still uses 8B/10B encoding and takes advantage of 4-channel CWDM to support 10Gb/s.<br />
In addition to PCS, PMA and PMD in LAN PHY, WAN PHY adds WAN interface sublayer (WIS) to support SONET line rate at 9.95328 Gb/s.<br />
10 Gigabit Ethernet has 7 interfaces:<br />
10GBase-SR-LAN-850nm, 10GBase-LR-LAN-1310nm, 10GBase-ER-LAN-1550nm, 10GBase-LX4-LAN-4X1310nm, 10GBase-SW-WAN-850nm, 10GBase-LW-WAN-1310nm and 10GBase-EW-WAN-1550nm.<br />
<ul><li>LAN/WAN PHY Sublayers</li>
</ul> 3 sublayers exist in LAN/WAN PHY: Physical coding sublayer, Physical medium attachment and Physical medium dependent.<br />
The PCS encodes and decodes data streams between the MAC and PHY layer. 3 categories of PCS are 10GBASE-R(64B/66B, 10.3 Gb/s, not SONET compatible), 10GBASE-X(8B/10B, WDM) and 10GBASE-W(64B/66B, 10 Gb/s, SONET compatible).<br />
PMA is an optional interface. 2 PMD interfaces are XAUI and XSBI. XAUI supports 4 SERDES transmit and 4 SERDES receive for 8B/10B encoding while XSBI requires more power and more pins.<br />
PMD is used to meet distance objective. 4 different PMD are defined: 850nm serial, up to 65 meters;<br />
1310 nm serial, upto 10 km, 1550 nm serial, upto 40 km, 1310 nm CWDM, upto 300 meters<br />
<br />
<div style="text-align: center;"> <a href="http://l3l2.blogspot.com/2010/12/ethernet-tutorial-2.html"> >>> NEXT</a></div><br />
<ol></ol><div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-8454324084287007505?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-64375759395610885962010-12-18T15:15:00.000-08:002010-12-21T22:05:16.239-08:002010-12-21T22:05:16.239-08:00Data Center, Cloud Computing and Switch/Router ResourcesSwitch/Router Architecture:<br />
<ul><li> <a href="http://www.ics.forth.gr/carv/psa.html">Packet Switch Architecture </a></li>
<li><a href="http://www.cs.virginia.edu/%7Ecs458/"> Internet Engineering @ Virginia University </a></li>
</ul><br />
Cloud Computing :<br />
<ul><li><a href="http://www.thecloudtutorial.com/">The Cloud Tutorial</a></li>
<li><a href="http://research.microsoft.com/en-us/labs/xcg/default.aspx">eXtreme Computing Group </a></li>
</ul><br />
All about Virtualization:<br />
<ul><li><a href="http://www.virtuatopia.com/index.php/Main_Page">Virtuatopia </a></li>
</ul><div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-6437575939561088596?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-35357887573809464682010-12-18T15:12:00.000-08:002010-12-19T00:18:56.977-08:002010-12-19T00:18:56.977-08:00Introduction to switch/routerFirst, let's think about postal system in real life. The system consists of branch office and main office. In branch office, there is one table which records each house hold's detailed address. In main post office, it have one table which records the postal code the corresponding branch office will process. When you want to deliver your mail, you will put the " to" and "from" address onto the envelop. Local postal man collect your mail and send to branch office. At branch office, staff will check whether the destination address of your mail belongs to local postal code. If yes, they will deliver your mail to the address directly; If not, they send your mail to main office. Main office will classify mails according to postal code of "to address", put them into different buckets and ship to next office which is responsible for processing mails with corresponding postal code. The peer main office will deliver those mails to branch office ; the peer branch office will deliver the mail to the local address if the destination address is one of household in its area.<br />
<br />
In fact, switch/router works like real postal system. Switch is like branch office; while router like main office. Switch/Router is the device which help the incoming packets find the right outgoing interface, which is closer to the destination of the packet, by checking look-up tables. Switch/router is the key components in all modern communication network such as wireless cellular network, telephone network, ATM network, IP data network and so on.<br />
<br />
<a name='more'></a><br />
In today's IP network, Ethernet switch and IP router are 2 important devices. Ethernet switch is responsible for switching the Ethernet packets from the incoming interface to the right outgoing interface. The outgoing interface may be the destination host directly, may be another switch which is closer to the destination, or may be one router who knows better about the destination and can help route the packets closer to their destination. To do the right decision, Ethernet switch depends on source and destination Ethernet MAC address, while IP router depends on the source and destination IP address.<br />
<br />
Switch and router are very similar in modern life. For chassis based switch/router, they are composed of chassis, Control engine, line cards, switch fabric, power supply and fan trays. Chassis is the container for all other components with back or middle planes. In one chassis, it could hold multiple line cards, switch fabrics, power supplies and fan trays. Control engine are the brain of modern switch/routing, and its responsibility is to learn network topology information from neighbors , construct a network picture of what it knows and program those knowledge into the look-up engine on the line cards. Line cards are composed of transceivers, PHY/EDCs, MACs, commercial or customized look-up engines, and optional switch fabric chips. Switch fabric connected all the line cards, and provide non-blocking packet delivery between line cards. Power supply provides power for the chassis, while fan tray holds the fans to cool all the electrical devices inside switch/routing.<div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-3535788757380946468?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-34467520897477916372010-12-18T14:15:00.000-08:002010-12-20T08:18:37.988-08:002010-12-20T08:18:37.988-08:00Introduction to switch fabricto do ...<div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-3446752089747791637?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-90511013523406173642010-12-18T14:14:00.001-08:002010-12-20T08:18:01.253-08:002010-12-20T08:18:01.253-08:00Intoduction to Network ProcessorTo do ....<div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-9051101352340617364?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-26944134115702288882010-12-18T14:13:00.001-08:002010-12-21T18:07:57.489-08:002010-12-21T18:07:57.489-08:00Introduction to MACMedium Access Control (MAC) is connected to PHY via Media Independent Interface (MII) toward network side. MII could be GMII, XGMII, XLGMII or CGMII, and it depends on the speed --- 1GbE, 10GbE, 40GbE or 100GbE. On the other end, MAC is connected to host via TX FIFO or RX FIFO. In most implementation, MAC consists of Reconciliation Sublayer, RX control, TX control, Receive FIFO, TX FIFO, Time stamping, Configuration and Statistics registers, and MDIO master. The RS is responsible for spraying packets onto the interface defined by MII and local/remote fault stuff.<br />
<a name='more'></a>The functionality of one general MAC is listed as following:<br />
<ol><li>Transit or receive Ethernet frames from PHY, detect or insert preamble and SFD, verify or calculate CRC-32 or FCS, terminate or insert optional padding, terminate pause or generate Xon / Xoff pause, extract or add IEEE1588 time stamp</li>
<li>Support configuring MAC address, frame length, half/full duplex, promiscuous mode</li>
<li>Support IEEE 802.1Q VLAN </li>
<li>Support statistics compatible with SNMP or RMON </li>
<li> Support MDIO master to access PHY </li>
</ol><div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-2694413411570228888?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-81240280669825000912010-12-18T14:05:00.000-08:002010-12-20T08:17:07.602-08:002010-12-20T08:17:07.602-08:00Introduction to PHYPHY is the physical layer of OSI network model. PHY consists of 3 sublayer generally -- PCS (Physical Coding Sublayer, PMA (Physical Medium Attachment Sublayer) and PMD (Physical Medium Dependent Sublayer). On one end, PHY is connected to optical transceivers or RJ45 jack directly; On the other end, PHY is connected to MAC.<br />
<br />
<a name='more'></a><br />
<br />
Inside PHY, there are some configuration and status registers, which could be accessed via MII interface by software programming. Mostly, MII is connect to MAC, and sometimes is hooked to host CPU like MIPS, ARM or PowerPC.<div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-8124028066982500091?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0tag:blogger.com,1999:blog-43233550506107058.post-80430946619265098502010-12-17T17:57:00.000-08:002010-12-20T08:16:28.622-08:002010-12-20T08:16:28.622-08:00Introduction to transceiversIn switch or router, the first component customers facing is optical transceiver, which is used to connected optical fibre to PHY or EDCs inside switch/router. Transceiver is the device to convert optical signal to electrical signal, which could be processed by modern switch/router. The following is one list of optical transceivers commonly used in networking industry.<br />
<ol><li>GBIC(10M/100M/1000M)</li>
<li>SFP/mini-GBIC(10M/100M/1000M)</li>
<li>XENPAK(10G)</li>
<li>X2(10G)</li>
<li>XFP(10G)</li>
<li>SFP+(10G)</li>
<li>QSFP(40G)</li>
<li>CXP(100G)</li>
<li>CFP(100G) </li>
</ol><div style="text-align: justify;"><br />
<a name='more'></a>Transceivers evolve with smaller size and higher speed. Most transceivers have the following features.<br />
<ul><li>have EEPROM accessible via I2C or MDIO </li>
<li>hot-pluggable</li>
<li>used with Ethernet or fibre channel. </li>
<li>support bi-directional communications, while others only support one direction -- either up link or down link only. </li>
<li>only support 1 speed, while others maybe support multiple speeds.</li>
<li> no PHY inside, while some have PHY embedded. </li>
</ul></div><div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/43233550506107058-8043094661926509850?l=l3l2.blogspot.com' alt='' /></div>l3l2nethttp://www.blogger.com/profile/10796771537034008648noreply@blogger.com0