Electrical computers and digital data processing systems: input/ – Intrasystem connection – Bus interface architecture
Reexamination Certificate
2000-09-29
2004-01-06
Myers, Paul R. (Department: 2181)
Electrical computers and digital data processing systems: input/
Intrasystem connection
Bus interface architecture
C710S317000, C709S239000
Reexamination Certificate
active
06675254
ABSTRACT:
BACKGROUND OF THE INVENTION
I. Field of the Invention
This invention relates generally to communication technology, and more particularly, to improving information transfer rates among all the cards in a rack mount system.
II. Background Information
Data processing systems typically experience data bottlenecks using busses such as those defined by the Industry Standard Architecture (“ISA”), the Extended Industry Standard Architecture (“EISA”), and the peripheral component interconnect (“PCI”). The PCI standard is a bus standard developed for personal computers (“PCs”) by the Intel Corporation that can transfer data between the,CPU and card peripherals at much faster rates than possible via the ISA bus. The mechanical, electrical, and operation characteristics for the current PCI bus standard may be found in PCI Local Bus Specification, Revision
2
.
1
, available from the PCI Special Interest Group.
PCI was originally designed for standardizing the interfaces available on chips to be used on PC compatible peripherals and was unique in that it utilized silicon. The PCI bus specification provided a processor-independent interface to add-in cards, also commonly referred to as expansion cards or adapters. Because of AC switching characteristic limitations, the PCI bus is typically limited in both information transfer rate and fan-out (number of add-in cards supported). The information transfer rate and fan-out in the PCI bus are interdependent, such that achieving an increase in one generally results in a decrease in the other.
Despite becoming the established interface, offering relatively high-speed data transfer, PCI lacked the higher density available from systems utilizing VersaModule Europe (“VME”), as only four cards could be supported within a system. CompactPCI was a solution this set of problems, given that it adopted the proven European form factor successfully utilized in VME systems.
CompactPCI is an adaptation of the PCI Specification 2.1 or later for industrial and/or embedded applications requiring a more robust mechanical form factor than desktop PCI. CompactPCI provides a system that is electrically compatible with the PCI Specification, allowing low cost PCI components to be utilized in a mechanical form factor suited for rugged environments. The mechanical, electrical, and operation characteristics for the current CompactPCI standard may be found in the CompactPCI Specification, PICMG 2.0R3.0, available from the PCI Industrial Computer Manufacturers Group (“PICMG”).
Unlike its desktop cousin, the CompactPCI comes in a rugged 3U (100 by 160 mm) or 6U (233 by 160 mm) VME packaging and uses a high quality 2 mm metric pin and socket connector that can be front loaded into a rack mount system. The CompactPCI has a 32 or 64-bit synchronous data bus, 32 bit address bus, 133 or 266 Mbytes/s data transfer rate, multiprocessing, and up to eight slot backplanes for add-in cards (the eight add-in cards interconnected together are known as a bus segment). Because of its extremely high bandwidth, the CompactPCI bus is particularly well suited for many high speed data communication applications such as switches. A hot swap feature is included in the CompactPCI specification which makes CompactPCI particularly well suited for the telecommunication industry.
Compared to the standard desktop PCI, Compact PCI supports twice as many PCI add-in cards (eight versus four) and offers a packaging scheme (i.e., the rack mount system) that is much better suited for use in industrial applications. The CompactPCI cards are designed for front loading and removal from a card cage within the rack mount system. The cards are firmly held in position by their connector, card guides on both sides, and a face plate which solidly screws into the card cage. Cards are mounted vertically allowing for natural forced air convection cooling.
A CompactPCI rack mount system (i.e., a system that includes add-in cards that conform to the CompactPCI form factor) supports the interconnection of CompactPCI form factor add-in cards within the system. The CompactPCI rack mount system may integrate a CompactPCI bus (i.e., a PCI bus integrated within the CompactPCI rack mount system). Information is transferred between the add-in cards connected to the CompactPCI bus at a rate of 1,056 MBit/s. The CompactPCI bus does not provide data redundancy, i.e., two sets of data is not transmitted to overcome errors in one of the sets of data.
The CompactPCI rack mount system may also integrate a H.110 TDM Telephony bus (the “H.110 bus”). The H.110 bus is known as the computer telephony specification or PICMG 2.5. The H.110 bus defines a single, industry-wide Computer Telephony Bus allowing the TDM interconnection of CompactPCI form factor telephony cards and peripherals within the rack mount system. The H.110 bus is not limited to only interconnecting a maximum of eight cards as is the case with the CompactPCI bus. The H.110 bus typically has a transfer rate of 256 MBit/s. In terms of switching bandwidth, the H.110 bus has a maximum capacity of 4096 bidirectional timeslots, each of the timeslots with a capacity 64 Kbit/s. However, if the peripheral card requires more timeslots than 4096 (e.g., cards supporting 3,072 simultaneous V.90 modems require 6,144 timeslots), then the H.100 bus is not adequate to support this bandwidth. As with the CompactPCI bus, the H.110 bus also does not provide data redundancy which is critical in some applications.
Deficiencies in the typical rack mount system may include the CompactPCI bus being slow (processes 1,056 MBit/s) for contemporary applications, and because of load considerations, the CompactPCI bus can only support up to eight cards (which is a relatively low fan-out) and does not provide data redundancy which may be critical for some applications. The H.110 TDM Telephony bus is also too slow (processes 256 MBit/s) thus not providing adequate bandwidth for contemporary applications and also does not provide data redundancy.
For the foregoing reasons, there is a need to connect cards in the rack mount system such that the connection is fast, supports a workable number of cards (i.e., has a high fan-out), and provides for data redundancy which is critical for some applications.
REFERENCES:
patent: 5161156 (1992-11-01), Baum et al.
patent: 5978881 (1999-11-01), Lebhar
patent: 6105122 (2000-08-01), Muller et al.
patent: 6112271 (2000-08-01), Lanus et al.
patent: 6134589 (2000-10-01), Hultgren
patent: 6247077 (2001-06-01), Muller et al.
patent: 6456498 (2002-09-01), Larson et al.
patent: 6473827 (2002-10-01), McMillen et al.
Savory, US 2001/00361178, Telecommunications switch with programmable call processing and real-time account management for switching and billing call all common informational protocols on a single switch and network, Nov. 1, 200.*
CompactPCI Specification Short Form, Sep. 2, 1997.
Intel Corporation
Kenyon & Kenyon
Myers Paul R.
Phan Raymond N
LandOfFree
System and method for mid-plane interconnect using switched... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for mid-plane interconnect using switched..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for mid-plane interconnect using switched... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3231268