A Compact Priority based Architecture Designed and Simulated for Data Sharing based on Reconfigurable Computing

Embed Size (px)

Citation preview

  • 7/30/2019 A Compact Priority based Architecture Designed and Simulated for Data Sharing based on Reconfigurable Computi

    1/4

    A Compact Priority based ArchitectureDesigned and Simulated for Data Sharing

    based on Reconfigurable ComputingBhavya Alankar and B K Kanaujia

    AbstractReconfigurable Computing devices are coming very strongly in the digital hardware systems due to the availability ofready to use resources, parallel logic operations and reconfigurable designs. The usage of Reconfigurable systems in real time do-main is also a very fruitful proposition as the FPGA devices are coming with processing cores for Real Time data processing. This

    paper mainly focuses on the application of Reconfigurable Computing systems in the data sharing domain and in order to depict thisapplication we have designed and simulated a priority based data sharing architecture through which we can interface two Proces-sors, which in turn can intercommunicate with each other. The main advantage of this architecture is that it is highly compact, easyto use and designed using a modular approach so that it could be easily implemented on Reconfigurable hardware. All the different

    modules along with the complete architecture is simulated using modelsim 6.0 and synthesized using Xilinx 7.1(iSE).

    Index TermsReconfigurable Computing, FPGA, Master-Slave processor.

    1 INTRODUCTION

    econfigurable computing based architectures offers aunique opportunity to address the memory accessand interfacing issues via customization. Improve-

    ments can be achieved by creating specialized architec-tures and this architecture is mainly designed to explorethose opportunities of Reconfigurable computing [1], [2],[3]. In our design FPGA Spartan-3 kit [7] is acting as aninteractive device through which two processors can in-tercommunicate with each other such that one processorhas been given priority over the other processor and their

    priority is been decided by the conflict resolving block.Initially this design is simulated for 8-bit processor but itcan be extended up to any number of bits.

    The organization of this paper is as follows: Section 2shows the design overview and the details of differentfunctional modules. Section 3 depicts the simulationwaveform of the architecture. Section 4 discuss the con-clusion and future work

    2 DESIGNOVERVIEW

    Thedesigned architecture is fully user friendly, flexible

    andsynchronous.

    One

    Processor

    acts

    as

    MASTER

    and

    the

    otherProcessoractslikeaSLAVEandcanintercommunicatewitheachother[4].Memoryprovidedviatwoseparateandfullysynchronousports.Masterwillalwayshave

    the priority over slave and we have usedmultiplexedaddress databus for themasterwhich canbe used forslaveaswell.Moreover in ourdesignwe haveused the signals of

    peripheralcomponentinterconnectbus[5]forthemasterprocessor to communicate and the slave processor cancommunicateviabusy,address,dataandreadwritesignalsasshowninFig1.,butdependingupontheapplicationyoucanmodifythem.Thedetaileddescriptionofeachoftheblocksdepicted

    inthe

    main

    block

    diagram

    along

    with

    their

    functional

    descriptionanddeviceutilizationsummary isgivenandit canbe seen that thearchitecture ishighly compactasthedeviceutilizationisquietlow.

    2.1 Conflict Resolving Block

    This block grants the control for accessing Dualportmemory to the requestingsystem.wearemainlyhavingtwo signals viz slave request (pr) and master request(pcirq)andthemainfunctioningofthisblockistoresolvethe conflictbetween thembymaking either slavebusy(probussy)ormasterbusy(pcibussy)aslow(seeTable1)

    TABLE1TRUTHTABLEOFCONFLICTRESOLVINGBLOCK

    pcirqmaster

    prslave

    pcibussymaster

    Probussyslave

    Low Low Low HighLow High Low HighHigh Low High LowHigh High High High

    Bhavya Alankar is a Assistant Professor, Hamdard University, New Delhi,India.

    B.K. Kanaujia is a Associate Profoessor, Ambedkar Institute of AdvancedCommunication Technologies & Research (AIACTR) Govt. of N.C.T..NewDelhi, India.

    R

    JOURNAL OF COMPUTING, VOLUME 5, ISSUE 4, APRIL 2013, ISSN (Online) 2151-9617

    https://sites.google.com/site/journalofcomputing

    WWW.JOURNALOFCOMPUTING.ORG 33

  • 7/30/2019 A Compact Priority based Architecture Designed and Simulated for Data Sharing based on Reconfigurable Computi

    2/4

    Fig.1.CompactArchitectureofDatasharingbetweenMasterSlave.

    Based on the information obtained from the abovetruth table, Two Boolean equations have been derivedand based on that equations, following digital design ofConflict resolving block has been obtained in Fig 2.

    Fig. 2.Digital design of conflict resolving block

    The device utilization summary of thisblock implementedontheSpartan3XC3S200FPGA(200Kgates) isshown inTable2.And It shows that thisblockwasdesignedinsuchawaysoastoutilizetheminimumnumber

    of

    resources

    on

    FPGA.

    TABLE2DEVICE UTILIZATION SUMMARY OF CONFLICT RE-

    SOLVING BLOCK

    Number ofslices

    1 out of 1920 0%

    Number of 4input LUTs

    1 out of 3840 0%

    Number ofbonded IOBs

    3 out of 173 1%

    2.2 Interactive Controller Blocks

    Whenboth(MasterandSlave)requestfortheaccessofmemory,TheConflictresolvingblockdrivesthebusyline of theMaster(pcibussy)while keeping the Slaveprocessor inawait stateuntil theMaster finishes itsoperation.When theMaster generates the address on

    address

    lines,

    data

    on

    data

    lines

    and

    write/read

    signal

    thenthecontrolblockcheckfor thebusy lineofMaster.When the controlblock receives thebusy line ofMaster driven to low, it allows the address on validaddresslines,dataonvaliddatalinesandwr_rdsignalasvalidwr_rdsignaltoaccessthememory.WhentheSlaveProcessorgeneratestheaddressonaddresslines,dataondata linesandwrite/readsignalthenthecontrolblockcheck for thebussy lineofSlaveProcessor.Whenthecontrolblockreceivesthebusylineof SlaveProcessor(probussy)driventolow itallowstheaddresson valid address lines, data on val id data lines andwr_rdsignalasvalidwr_rdsignaltoaccessthememoryasshowninFig3.

    Fig.3.InteractiveControllerBlock

    Now its clear that themain function of the Interactivecontrollerblock is to just select or decidewhether theMasterorSlaveshouldaccessthememoryonthebasisoftheoutputofConflictresolvingblock,Sothisblockisjusta combination of the Conflict resolving block and the

    memory

    read

    and

    write

    block.

    ThedeviceutilizationsummaryoftheInteractiveController block implemented on the Spartan3 XC3S200FPGA (200Kgates) isshown inTable3.As thisblock isthecombinationofmainlythemuxes.Itshowsthenumberofslicesandotherresourcesavailableutilized in implementingthedesignonFPGA.

    JOURNAL OF COMPUTING, VOLUME 5, ISSUE 4, APRIL 2013, ISSN (Online) 2151-9617

    https://sites.google.com/site/journalofcomputing

    WWW.JOURNALOFCOMPUTING.ORG 34

  • 7/30/2019 A Compact Priority based Architecture Designed and Simulated for Data Sharing based on Reconfigurable Computi

    3/4

    TABLE3DEVICEUTILIZATIONSUMMARYOFINTERACTIVE

    CONTROLLERBLOCK

    2.3 Memory block

    Itisoneofthemainblockandasitisclearfromthenamethatthemainpurposeofthisblockistofulfillthestoragerequirementofboth theMasterand theSlaveProcessor.

    Thisstorage

    core

    is

    adual

    port

    memory

    of

    64

    ksize,

    fully

    synchronous such that the data is tobewritten on therisingedgeof theclock. It is thecentralblockwhichdependsontheoutputofalltheblockssuchthatfirstofalltheConflict resolvingblockwilldecidewhetherMasterorSlaveProcessorwillaccessthememoryanddependingon theoutputof that the controllerblockwillallow thevalidsignalstoaccessthememoryasshowninFig.4.Thedeviceutilization summary of thememoryblock

    implemented on the Spartan3 XC3S200 FPGA (200 Kgates) isshown inTable4. Itshowsthenumberofslicesand other resources available utilized in implementingthedesignonFPGA.

    Fig. 4. Memory Block and Its Connectivity

    TABLE4DEVICEUTILIZATIONSUMMARYOFMEMORY

    BLOCK

    Numberofslices 9outof1920 0%

    Numberofsliceflip

    flops

    16outof3840 0%

    Numberof4inputLUTs 3outof3840 0%

    NumberofbondedIOBs 49outof173 28%

    NumberofBRAMs 1outof12 8%

    NumberofGCLKs 2outof8 25%

    3. SIMULATION RESULTS

    3.1MasterContinuouswriteandSlavecontinuousreadcycle:In this the Master take the control from the Conflictresolving block by driving its bus request (pcirq) to accessthe memory. Then it broadcasts the address and data onaddress_data line such that first the address is send andby driving frame# as low and irdy# and trdy# drivinghigh the address is latched and then the data is send andby driving frame# high and driving irdy# and trdy# lowthe data is latched and in the mean time thepciwr(Master write) signal is also made low such that thedata is written in the memory.After the data is written into the memory the data can be continuously read by theSlave processor by just giving the address in the read ad-dress line and by applying read clock and in the meantime the pw(Slave write) signal should also be high andthe data ouput will be coming on the Slave processordata (pd) output line, Fig 5.

    Fig. 5. Master Write and Slave read cycle

    Numberof

    slices

    1out

    of

    1920

    0%

    Numberof4inputLUTs 1outof3840 0%

    NumberofbondedIOBs 4outof173 1%

    NumberofGCLKs 1outof8 15%

    JOURNAL OF COMPUTING, VOLUME 5, ISSUE 4, APRIL 2013, ISSN (Online) 2151-9617

    https://sites.google.com/site/journalofcomputing

    WWW.JOURNALOFCOMPUTING.ORG 35

  • 7/30/2019 A Compact Priority based Architecture Designed and Simulated for Data Sharing based on Reconfigurable Computi

    4/4

    3.2SlaveContinuouswrite andSlave continuous readcycle:In this the Slave processor take the control from theConflict resolving block by driving its bus request (pr) toaccess the memory. Then it broadcasts the address onaddresslines (pa [0:7]) and data on datalines (pd [0:7])then it drives its write/read signal (pw) to low after thatit drives its chip enable signal (pe) to low, which causesthe data to be written to memory of the specified ad-dress.For reading the data from memory the Slave pro-cessor broadcasts the address on addresslines (pa [0:7])then it drives its write/read signal (pw) to high after thatit drives its enable signal (pe) to low, which causes thedata to be read from the memory corresponding to thespecified address,asshowninFig 6.

    Fig. 6. Slave Write and Slave read cycle

    3.3 Master Continuous write and Master continuousread cycle: In this the Master take the control from theConflict resolving block by driving its bus request (pcirq)to access the memory. Then it broadcasts the address anddata on address_data line such that first the address is

    send and by driving frame# as low and irdy# and trdy#driving high ,the address is latched and then the data issend and by driving frame# high and driving irdy# andtrdy# low ,the data is latched and in the mean time thepciwr(Master write) signal is also made low such that thedata is written in the memory. After the data is written into the memory the data can be continuously read by justgiving the address in the same way on the address_dataline and afterwards just driving the pciwr signal as highsuch that the data will appear on the output line(pcida),Fig 7.

    Fig. 7. Master Write and Master read cycle

    4 CONCLUSION

    TheperformanceofanyReconfigurableComputing systemmainlydependscriticallyuponitsflexibilityandeaseof interfacingwithotherhardware architectures. In thispaperwe have described an architecturewhich couldeasilybeusedfor intercommunicationbetweentheprocessors

    in

    prioritized

    manner

    .We

    have

    also

    described

    clearlythearchitectureandworkingofdifferentmodulesalongwiththeirdeviceutilizationsummary.Thearchitectureisfullysynthesizedandsimulatedforinterprocessorcommunication. The ongoingwork is the implementation of this architecture using FPGA kit alongwith themicrocontrollers which could easily depicts the interprocessorcommunicationinrealtime,asthisarchitectureisbasicallydesigned to support the advances inReconfigurableComputing.

    REFERENCES

    [1]Joa

    O

    m,.

    P.

    Cardoso,

    Pedro

    C.

    Diniz

    and

    Markus

    WeinHardt,

    Compiling for Reconfigurable Computing: A Survey, ACMComputingSurveys,Vol.42,No.4,Article13,Publicationdate:

    June2010.

    [2] Katherine ComptonandScott Hauck, ReconfigurableCom

    puting:ASurvey of Systems and Software,ACMComputing

    Surveys,Vol.34,No.2,June2002,pp.171210.

    [3] T.J.Todman,G.A.Constantinides,S.J.E.Wilton,O.Mencer,W.

    Luk and P.Y.K. Cheung Reconfigurable Comput

    ing:architectures and designmethods.published in IEE Proc.

    Comput.Digit.Tech.,Vol.152,No.2,March2005

    [4] Banerjee,T.P,konar,chaudhary Highspeed communication

    system developed using FPGA based CAM implementation

    publishedinICETET2009

    [5] PCIbusmanualwebsite

    :http://www.pciadda.com/manual/manual1.html[6] Xilinxuser

    guide:http://www.xilinx.com/itp/xilinx10/books/docs/xst/xst.pd

    f

    [7] FPGASpartan3 website:

    http://www.xilinx.com/support/documentation/boards_and_kit

    s/ug230.pdf

    Bhavya Alankaris a Assitant Professor in Hamdard University, NewDelhi. He has acquired his Master degree in VLSI design. His mainresearch interests areReconfigurable computing, VLSI design andSoftware Engineering.He is having years of research experience inVLSI design and published numerous research papers and conduct-ed workshops related to his research area.

    B K Kanaujia is a Associte Professor in Ambedkar Institute ofAdvancedCommunicationsTechnologies& Research (AIACTR), Govt.of N.C.T.New Delhi, India. He did his M.Tech. & Ph.D. from Electron-ics Engineering Department of Banaras Hindu University Varanasi,His keen research interest are in design and Modeling ofMicrostrip Antenna, Dielectric Resonator Antenna, Left handedMetamaterial Microstrip Antenna, Shorted Microstrip Antenna Wire-less Communication, Wireless Communication, Microwave Engi-neering,Reconfigurable computing and VLSI design etc.Till date hehas been credited to publish more than 45 research papers in peer-reviewed journals and International/national conferences.

    JOURNAL OF COMPUTING, VOLUME 5, ISSUE 4, APRIL 2013, ISSN (Online) 2151-9617

    https://sites.google.com/site/journalofcomputing

    WWW.JOURNALOFCOMPUTING.ORG 36