Who We Are - PowerPoint PPT Presentation

1 / 18
About This Presentation
Title:

Who We Are

Description:

Fibre Channel Preprocessor for Logic Analyzers ... JTAG BST lines. 32 bit. 32 bit. 4. 1. FEE - SIU Interface. DIU -RORC Interface ... – PowerPoint PPT presentation

Number of Views:39
Avg rating:3.0/5.0
Slides: 19
Provided by: kissti
Category:
Tags: bst

less

Transcript and Presenter's Notes

Title: Who We Are


1
Who We Are?
Detector Building Group of KFKI-RMKI(Research
Institute for Particle and Nuclear Physics),
Budapest, HUNGARY
2
Our Results
  • Previous projects
  • Designing Fibre Channel test equipments
    (1994-1998)
  • Portable, 266 Mb/s Fibre Channel Tester
  • Fibre Channel Preprocessor for Logic Analyzers
  • Designing high-speed data transmission interfaces
    for CERN detectors (1996 - ...)
  • Protocol design and verification
  • FC, GbE, physical layer components
  • Hardware design
  • Software linux drivers, test programs, program
    library

3
S-LINK (CERN)
  • S-LINK interface cards for CERN (mostly used at
    ATLAS)
  • simple, unidirectional link interface
  • first successful version was designed by
    KFKI-RMKI
  • newer versions, now 2.5 Gb/s
  • Also used at MPI, Garching and at several HEP and
    other scientific institutes all over the world

4
ALICE Detector Data Link (DDL)
5
DDL
  • Detector Data Link (DDL) for the ALICE detector
    at CERN
  • 2.5 Gb/s, duplex link
  • advanced featues
  • Test devices for DDL (Front-end emulator, DDL
    link emulator, etc.)
  • PCI Read-Out Receiver Card (RORC) for DDL
  • 1st version 33 MHz, 32-bit PCI card
  • 2nd version 66 MHz, 64-bit PCI with 2
    integrated DDL interfaces

6
DDL Interfaces
DIU -RORC Interface
FEE - SIU Interface
Read-out Receiver Card (RORC) 66 MHz 64-bit PCI
DDL Destination Interface Unit (DIU)
Front-end Electronics
DDL Source Interface Unit (SIU)
Front-end Bus 32 bit
32 bit
4
T A P
optical cable, max 350 m
1
32 bit
JTAG BST lines
DDL
7
DDL Features
  • Interface
  • Full duplex 32-bit data path on the destination
    interface (DIU card)
  • Half duplex 32-bit data path on the source
    interface (SIU card)
  • Full duplex flow control (XON/XOFF)
  • Interface clock up to 66 MHz (easy integration
    with PCI 66)
  • 264 MB/s peak data rate, 240 MB/s sustained
    bandwidth (max.)
  • Implementation
  • Duplex LC optical link up to 300 m
  • 2x FC or 2x GbE physical layer components
  • Small Form Factor Pluggable (SFP) optical
    transceivers
  • Bit error rate lt 10 -12
  • Robust error detection very low undetected bit
    error rate lt 10-40
  • Automatic link synchronization and management
  • Extras
  • Stand-by support (low power consumption)
  • In-system reconfiguration /  Remote system
    upgrade
  • Monitoring of the aging of laser diode of optical
    transceivers

8
Outlook to PCI Express
  • PCI Express (formerly 3GIO)
  • a third generation high performance I/O bus
  • (1st generation ISA, EISA, VESA, 2nd
    generation PCI, PCI-X)
  • PCI evolution (PCI, PCI-X, PCI-Express)
  • PCI Express is software compatible to PCI and
    PCI-X

9
PCI Evolution
  • PCI
  • multi-drop parallel bus
  • conventional PCI 33 MHz, 32-bit, max 4-5 card
    slots per bus
  • newer versions 66 MHz, 64-bit, 1 (max 2) card
    slots per bus (!)
  • PCI-X
  • parallel bus, backward compatible (hw and sw)
  • 66 MHz , 32/64-bit, max 4-5 card slots per bus
  • 133 MHz, 32/64-bit, max 1-2 card slots per bus
  • 266, 532 MHz versions max 1 card slot per bus
    (!)
  • Buses can be bridged to each other (complex,
    expensive)
  • PCI-XP
  • It is still a local /IO bus, a PCI bus, but the
    connections between devices are serial,
    point-to-point interconnections
  • devices are interconnected via switch(es)
  • large number of devices can be interconnected
  • highly scalable, hot-plug, hot-swap, QoS, etc.

10
Limitations of PCI and PCI-X
  • Conventional 33 MHz PCI system
  • low bandwith to nowadays needs
  • 66 MHz, 133 MHz (PCI-X), ...
  • only few devices can be interconnected (only 1 or
    two) on a single bus because of the strict
    electrical load and timing constraints
  • Further limitations of PCI architecture
  • inefficient solutions in
  • data transfer cycles (wait states)
  • accessing of system memory
  • interrupt handling
  • error handling

11
PCI Express
  • However, the basic problems of a parallel bus
    system(electrical load and timing constraints,
    lack of hot-pluggability, lack of scalability,
    etc.) can be solved only by a complete redesign
    of the architecture.
  • This resulted in PCI Express.

12
The PCI Express Link
  • Serial point-to-point interconnect between two
    devices
  • 1x, 2x, 4x, 8x, 12x, 16x, 32x type links
  • 1, 2, 4, ... 32 bidirectional signal pairs
    (lanes)
  • 1 lane 2.5 Gb/s now, up to 10 Gb/s later
  • Low voltage, differential signaling (LVDS)
  • AC coupled
  • Data is encoded 8B/10B

a PCI Express link
Device A
Device B (e.g. a switch)
1 to 32 lanes
13
Bandwith
  • Scalable more lanes /link higher bandwith
  • 2.5 Gb/s per lane, 8B/10B encoding
  • Simultaneous traffic in both directions
  • 1x type link
  • 500 MByte/s aggregate bandwith
  • 250 MByte/s per direction
  • 32x type link
  • 16 GByte/s aggregate bandwith
  • (8 GByte/s per direction)

14
Backward compatibility
  • Supports familiar PCI transactions such as memory
    read/write, IO read/write and configuration
    read/write
  • Same memory, IO and configuration address space
    as in PCI and PCI-X
  • Existing OSs and driver software will run in a
    PCI Express system without any modifications
  • Supports chip-to-chip interconnect and
    board-to-board interconnect via cards and
    connectors similar to the present PCI systems
  • PCI Express motherboard will have a similar form
    factor to existing ATX motherboards in PCs

15
Improvements, Benefits
  • Fast, highly scalable, serial point-to-point
    interconnect between two devices
  • bytes striped accross the lanes ? more lanes
    per link faster transmission
  • Packet-based communication protocol
  • Packets are transmitted serially
  • CRC embedded in each packet (auto retry link
    level error correction)
  • Buffer-to buffer (link level) flow control ? no
    need the packet retry
  • Message Signaled Interrupt (MSI) architecture
  • No side-band signals
  • hot-plug, error handling, interrupt signaling and
    else are accopmlished in-band
  • Multiple devices are interconnected via switches
  • Large number of devices can be connected together
    in a system
  • Much fewer pins per device package
  • Reduces chip and board design cost and design
    complexity

16
Outside the box?
  • Initial focus of usage inside the box
  • It is expected that later it will also be used
    outside the box for I/O expansion

PCI-XP links
PCI-XP links
Add-in card
Add-in card
switch
switch
opticalcable
connector
PCI-XP link
17
Mechanical Form Factors
  • Connector and daughter card form factors are
    under specification
  • The main type of connectors is very similar to
    the present PCI card edge connector
  • A 1x type card can be inserted in a x4 type slot,
    a.s.o.
  • Like with PCI, there will be other form factors
  • Server I/O module
  • Mini PCI Express card and connector (e.g. for
    notebooks)
  • mezzanine type card
  • NEWCARD (will replace CardBus PC card), etc.

18
Compete or complete?
  • PCI Express will coexist with PCI / PCI-X in the
    same system
  • PCI Express intends to replace AGP (graphics card
    if), but will not replace Serial ATA (hard disk,
    CD, etc)
  • HyperTransport
  • onboard chip-to-chip interconnect
  • said to be complementary to PCI Express
  • RapidIO
  • as an onboard chip-to-chip interconnect
    complementary to PCI-Express
  • as a system interconnect a competitive
    technology
  • 1394b, USB2.0, Fibre Channel, Gb Ethernet,
    Infinband, etc.
  • PCI Express can be a system interconnect that
    bridges these technologies
  • Will it be?
Write a Comment
User Comments (0)
About PowerShow.com