Friday, December 24, 2010

802.1p vs 802.1Q

802.1p 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.
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 .

Tuesday, December 21, 2010

Gb/s, GB/s and GT/s

In 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.
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.

Terminal Server

Terminal 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.

19 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.

When buying chassis based switch/routers or servers, good to buy those standard chassis. When looking into
the data sheets or production information next time,  you know what those terminology means.

23 inch or 24 in rack also exist in industry, but not as popular as 19 inch rack chassis.

Monday, December 20, 2010

Technology Websites

  1. EE Times
  2. NullModerm
  3. Light Reading
  4. Javvin 
  5. Tech Crunch
  6. Design & Resuse

BPS (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

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) .

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.

Sunday, December 19, 2010

Ethernet Tutorial 2

  1.  Ethernet Frames Format
  2. Ethernet Frames
    Field DIX Frame IEEE 802.3 Frame
    Preamble    8 bytes    7 bytes
    Start Frame Delimiter       1 bytes
    Destination Address    6 bytes    6 bytes
    Source Address   6 bytes    6 bytes
    Length      2 bytes
    Type    2 bytes   
    Data  1500 bytes     Yes
    CRC/FCS  4 bytes    4 bytes
    Total  1518 bytes    1518bytes
    Note: 100/1000Mbs Ethernet Systems do not need preamble or start frame delimiter fields.

Introduction to Cisco's Nexus Switches

While 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.
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
is a list of Nexus series switches. You could see its portfolio is the richest in industry.

Saturday, December 18, 2010

Introduction to Arista EOS Switches

Arista 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.

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.

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.

Ethernet Tutorial 1

One 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 here. You could also
check this link from England.
  1. Ethernet History
  2. 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.
  3. Ethernet Standards
  4. 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 802.3 .
  5.  Ethernet Technologies
  • Fast Ethernet 
         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.

Data Center, Cloud Computing and Switch/Router Resources

Switch/Router Architecture:

Cloud Computing :

All about Virtualization:

Introduction to switch/router

First, 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.

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.

Introduction to switch fabric

to do ...

Intoduction to Network Processor

To do ....

Introduction to MAC

Medium 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.

Introduction to PHY

PHY 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.

Friday, December 17, 2010

Introduction to transceivers

In 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.
  1. GBIC(10M/100M/1000M)
  2. SFP/mini-GBIC(10M/100M/1000M)
  3. XENPAK(10G)
  4. X2(10G)
  5. XFP(10G)
  6. SFP+(10G)
  7. QSFP(40G)
  8. CXP(100G)
  9. CFP(100G)