LZ4相关.rar

  • xiaouong
    了解作者
  • WINDOWS
    开发工具
  • 10.4MB
    文件大小
  • rar
    文件格式
  • 0
    收藏次数
  • 1 积分
    下载积分
  • 0
    下载次数
  • 2020-09-28 13:54
    上传日期
Lossless compression algorithm LZ4 related papers
LZ4相关.rar
  • LZ4相关
  • Hardware-accelerated Fast Lossless Compression Based on LZ4 Algorithm.pdf
    848.5KB
  • LZ4 Compression Algorithm on FPGA.pdf
    344.7KB
  • Data Compression Device Based on Modified LZ4 Algorithm .pdf
    1.7MB
  • Design of hardware accelerator for Lempel-Ziv 4 (LZ4) compression.pdf
    1.7MB
  • 基于FPGA的LZ4无损压缩算法优化设计_顾巍.caj
    8.1MB
内容介绍
<html xmlns="http://www.w3.org/1999/xhtml"> <head> <meta charset="utf-8"> <meta name="generator" content="pdf2htmlEX"> <meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1"> <link rel="stylesheet" href="https://static.pudn.com/base/css/base.min.css"> <link rel="stylesheet" href="https://static.pudn.com/base/css/fancy.min.css"> <link rel="stylesheet" href="https://static.pudn.com/prod/directory_preview_static/62509d696caf5961920d964e/raw.css"> <script src="https://static.pudn.com/base/js/compatibility.min.js"></script> <script src="https://static.pudn.com/base/js/pdf2htmlEX.min.js"></script> <script> try{ pdf2htmlEX.defaultViewer = new pdf2htmlEX.Viewer({}); }catch(e){} </script> <title></title> </head> <body> <div id="sidebar" style="display: none"> <div id="outline"> </div> </div> <div id="pf1" class="pf w0 h0" data-page-no="1"><div class="pc pc1 w0 h0"><img class="bi x0 y0 w1 h1" alt="" src="https://static.pudn.com/prod/directory_preview_static/62509d696caf5961920d964e/bg1.jpg"><div class="t m0 x1 h2 y1 ff1 fs0 fc0 sc0 ls0 ws0">110<span class="_ _0"> </span>IEEE<span class="_ _1"> </span>TRANSA<span class="_ _2"></span>CTIONS<span class="_ _1"> </span>ON<span class="_ _1"> </span>CONSUMER<span class="_ _1"> </span>ELECTR<span class="_ _2"></span>ONICS,<span class="_ _1"> </span>V<span class="_ _2"></span>OL.<span class="_ _1"> </span>64,<span class="_ _1"> </span>NO.<span class="_ _1"> </span>1,<span class="_ _1"> </span>FEBR<span class="_ _2"></span>U<span class="_ _2"></span>AR<span class="_ _2"></span>Y<span class="_ _1"> </span>2018</div><div class="t m0 x2 h3 y2 ff1 fs1 fc0 sc0 ls1 ws0">Data<span class="_ _3"> </span>Compression<span class="_ _3"> </span>De<span class="_ _4"></span>vice<span class="_ _3"> </span>Based<span class="_ _3"> </span>on</div><div class="t m0 x3 h3 y3 ff1 fs1 fc0 sc0 ls1 ws0">Modi&#64257;ed<span class="_ _3"> </span>LZ4<span class="_ _3"> </span>Algorithm</div><div class="t m0 x4 h4 y4 ff1 fs2 fc0 sc0 ls1 ws0">W<span class="_ _5"></span>eiqiang<span class="_ _6"> </span>Liu,<span class="_ _6"> </span><span class="ff2 ls2">Senior<span class="_ _6"> </span>Member<span class="_ _7"></span>,<span class="_ _6"> </span>IEEE<span class="ff1 ls3">,<span class="_ _6"> </span>Faqiang<span class="_ _6"> </span>Mei,<span class="_ _6"> </span>Chenghua<span class="_ _6"> </span>W<span class="_ _5"></span>ang,<span class="_ _6"> </span>Maire<span class="_ _6"> </span>O&#8217;Neill,<span class="_ _6"> </span><span class="ff2 ls2">Senior<span class="_ _6"> </span>Member<span class="_ _7"></span>,<span class="_ _6"> </span>IEEE<span class="_ _8"></span><span class="ff1 ls1">,</span></span></span></span></div><div class="t m0 x5 h4 y5 ff1 fs2 fc0 sc0 ls1 ws0">and<span class="_ _6"> </span>Earl<span class="_ _6"> </span>E.<span class="_ _6"> </span>Swartzlander<span class="_ _4"></span>,<span class="_ _6"> </span>Jr<span class="_ _4"></span>.,<span class="_ _6"> </span><span class="ff2 ls4">Life<span class="_ _6"> </span>F<span class="_ _5"></span>ellow<span class="_ _5"></span>,<span class="_ _6"> </span>IEEE</span></div><div class="t m0 x6 h5 y6 ff3 fs3 fc0 sc0 ls5 ws0">Abstract<span class="ff4 ls1">&#8212;Data<span class="_ _9"> </span>compression<span class="_ _9"> </span>is<span class="_ _9"> </span>commonly<span class="_ _9"> </span>used<span class="_ _a"> </span>in<span class="_ _9"> </span><span class="fs4 ls6">NAND<span class="_ _a"> </span></span><span class="ls7">&#64258;ash-</span></span></div><div class="t m0 x1 h5 y7 ff4 fs3 fc0 sc0 ls1 ws0">based<span class="_ _6"> </span>solid<span class="_ _b"> </span>state<span class="_ _b"> </span>drives<span class="_ _6"> </span>(SSDs)<span class="_ _b"> </span>to<span class="_ _6"> </span>incr<span class="_ _4"></span>ease<span class="_ _6"> </span>their<span class="_ _b"> </span>storage<span class="_ _6"> </span>perfor<span class="_ _4"></span>-</div><div class="t m0 x1 h5 y8 ff4 fs3 fc0 sc0 ls1 ws0">mance<span class="_"> </span>and<span class="_ _1"> </span>lifetime<span class="_ _1"> </span>as<span class="_ _1"> </span>it<span class="_ _1"> </span>can<span class="_"> </span>reduce<span class="_ _1"> </span>the<span class="_"> </span>amount<span class="_ _1"> </span>of<span class="_ _1"> </span>data<span class="_ _1"> </span>written<span class="_ _1"> </span>to</div><div class="t m0 x1 h5 y9 ff4 fs3 fc0 sc0 ls8 ws0">and<span class="_ _6"> </span>read<span class="_ _6"> </span>from</div><div class="t m0 x7 h5 ya ff4 fs4 fc0 sc0 ls6 ws0">NAND<span class="_ _6"> </span><span class="fs3 ls1">&#64258;ash<span class="_ _6"> </span>memory<span class="_ _5"></span>.<span class="_ _c"> </span>Software-based<span class="_ _6"> </span>data<span class="_ _6"> </span>com-</span></div><div class="t m0 x1 h5 yb ff4 fs3 fc0 sc0 ls1 ws0">pression<span class="_ _d"> </span>reduces<span class="_ _d"> </span>SSD<span class="_ _d"> </span>performance<span class="_ _d"> </span>signi&#64257;cantly<span class="_ _e"> </span>and,<span class="_ _d"> </span>as<span class="_ _e"> </span>such,</div><div class="t m0 x1 h5 yc ff4 fs3 fc0 sc0 ls1 ws0">hardwar<span class="_ _2"></span>e-based<span class="_ _f"> </span>data<span class="_ _f"> </span>compression<span class="_ _e"> </span>designs<span class="_ _f"> </span>are<span class="_ _f"> </span>requir<span class="_ _2"></span>ed.<span class="_ _f"> </span>This</div><div class="t m0 x1 h5 yd ff4 fs3 fc0 sc0 ls5 ws0">paper<span class="_ _f"> </span>studies<span class="_ _f"> </span>the<span class="_ _f"> </span>latest<span class="_ _f"> </span>lossless<span class="_ _10"> </span>data<span class="_ _f"> </span>compression<span class="_ _f"> </span>algorithm,</div><div class="t m0 x1 h5 ye ff4 fs3 fc0 sc0 ls5 ws0">i.e.,<span class="_ _b"> </span>the<span class="_ _6"> </span>Lempel-Zi<span class="_ _2"></span>v<span class="_ _b"> </span>(LZ)4<span class="_ _b"> </span>algorithm<span class="_ _6"> </span>which<span class="_ _b"> </span>is<span class="_ _b"> </span>one<span class="_ _b"> </span>of<span class="_ _6"> </span>the<span class="_ _b"> </span>fastest</div><div class="t m0 x1 h5 yf ff4 fs3 fc0 sc0 ls1 ws0">compression<span class="_ _c"> </span>algorithms<span class="_ _d"> </span>reported<span class="_ _d"> </span>to<span class="_ _c"> </span>date.<span class="_ _d"> </span>A<span class="_ _d"> </span>data<span class="_ _c"> </span>compression</div><div class="t m0 x1 h5 y10 ff4 fs3 fc0 sc0 ls5 ws0">FPGA<span class="_"> </span>pr<span class="_ _2"></span>ototype<span class="_"> </span>based<span class="_ _11"> </span>on<span class="_"> </span>the<span class="_ _11"> </span>LZ4<span class="_"> </span>lossless<span class="_ _11"> </span>compression<span class="_"> </span>algorithm</div><div class="t m0 x1 h5 y11 ff4 fs3 fc0 sc0 ls1 ws0">is<span class="_ _6"> </span>studied.<span class="_ _b"> </span>The<span class="_ _6"> </span>original<span class="_ _b"> </span>LZ4<span class="_ _6"> </span>compr<span class="_ _4"></span>ession<span class="_ _6"> </span>algorithm<span class="_ _b"> </span>is<span class="_ _6"> </span>modi&#64257;ed</div><div class="t m0 x1 h5 y12 ff4 fs3 fc0 sc0 ls8 ws0">for<span class="_ _a"> </span>real-time<span class="_ _9"> </span>hardwar<span class="_ _2"></span>e<span class="_ _9"> </span>implementation.<span class="_ _9"> </span>T<span class="_ _4"></span>wo<span class="_ _9"> </span>hardwar<span class="_ _2"></span>e<span class="_ _9"> </span>architec-</div><div class="t m0 x1 h5 y13 ff4 fs3 fc0 sc0 ls5 ws0">tures<span class="_ _9"> </span>of<span class="_ _b"> </span>the<span class="_ _9"> </span>modi&#64257;ed<span class="_ _b"> </span>LZ4<span class="_ _b"> </span>algorithm<span class="_ _b"> </span>(MLZ4)<span class="_ _9"> </span>are<span class="_ _9"> </span>proposed<span class="_ _b"> </span>with</div><div class="t m0 x1 h5 y14 ff4 fs3 fc0 sc0 ls1 ws0">both<span class="_ _d"> </span>compressors<span class="_ _c"> </span>and<span class="_ _d"> </span>decompressors,<span class="_ _d"> </span>which<span class="_ _d"> </span>ar<span class="_ _2"></span>e<span class="_ _d"> </span>implemented</div><div class="t m0 x1 h5 y15 ff4 fs3 fc0 sc0 ls5 ws0">on<span class="_ _d"> </span>an<span class="_ _c"> </span>FPGA<span class="_ _d"> </span>e<span class="_ _2"></span>valuation<span class="_ _c"> </span>kit.<span class="_ _d"> </span>The<span class="_ _12"> </span>implementation<span class="_ _12"> </span>results<span class="_ _c"> </span>show</div><div class="t m0 x1 h5 y16 ff4 fs3 fc0 sc0 ls1 ws0">that<span class="_ _c"> </span>the<span class="_ _12"> </span>proposed<span class="_ _c"> </span>compressor<span class="_ _c"> </span>architecture<span class="_ _c"> </span>can<span class="_ _12"> </span>achieve<span class="_ _c"> </span>a<span class="_ _c"> </span>high</div><div class="t m0 x1 h5 y17 ff4 fs3 fc0 sc0 ls1 ws0">throughput<span class="_ _6"> </span>of<span class="_ _6"> </span>up<span class="_ _c"> </span>to<span class="_ _6"> </span>1.92<span class="_ _c"> </span>Gb/s<span class="_ _6"> </span>with<span class="_ _c"> </span>a<span class="_ _6"> </span>compression<span class="_ _6"> </span>ratio<span class="_ _c"> </span>of<span class="_ _6"> </span>up</div><div class="t m0 x1 h5 y18 ff4 fs3 fc0 sc0 ls1 ws0">to<span class="_ _b"> </span>2.05,<span class="_ _b"> </span>which<span class="_ _6"> </span>is<span class="_ _b"> </span>higher<span class="_ _b"> </span>than<span class="_ _6"> </span>all<span class="_ _9"> </span>previous<span class="_ _b"> </span>LZ<span class="_ _b"> </span>algorithm<span class="_ _b"> </span>designs</div><div class="t m0 x1 h5 y19 ff4 fs3 fc0 sc0 ls1 ws0">implemented<span class="_ _9"> </span>on<span class="_ _9"> </span>FPGAs.<span class="_ _9"> </span>The<span class="_ _9"> </span>compression<span class="_ _9"> </span>device<span class="_ _9"> </span>can<span class="_ _9"> </span>be<span class="_ _9"> </span>used<span class="_ _b"> </span>in</div><div class="t m0 x1 h5 y1a ff4 fs3 fc0 sc0 ls1 ws0">high-end<span class="_ _1"> </span>SSDs<span class="_ _a"> </span>to<span class="_ _1"> </span>further<span class="_ _a"> </span>increase<span class="_"> </span>their<span class="_ _a"> </span>storage<span class="_ _1"> </span>performance<span class="_ _1"> </span>and</div><div class="t m0 x1 h5 y1b ff4 fs3 fc0 sc0 ls1 ws0">lifetime.</div><div class="t m0 x6 h5 y1c ff3 fs3 fc0 sc0 ls1 ws0">Index<span class="_ _c"> </span>T<span class="_ _5"></span>erms<span class="ff4 ls5">&#8212;FPGA,<span class="_ _12"> </span>lossless<span class="_ _12"> </span>compression,<span class="_ _c"> </span>Lempel-Ziv<span class="_ _c"> </span>(LZ)</span></div><div class="t m0 x1 h5 y1d ff4 fs3 fc0 sc0 ls5 ws0">algorithms,<span class="_ _b"> </span>LZ4,<span class="_ _b"> </span>solid-state<span class="_ _b"> </span>drives<span class="_ _9"> </span>(SSDs).</div><div class="t m0 x8 h6 y1e ff1 fs5 fc0 sc0 ls9 ws0">I.<span class="_ _e"> </span>I<span class="fs6 lsa">NTR<span class="_ _4"></span>ODUCTION</span></div><div class="t m0 x1 h7 y1f ff4 fs7 fc0 sc0 ls1 ws0">S</div><div class="t m0 x9 h6 y20 ff1 fs5 fc0 sc0 lsb ws0">OLID-ST<span class="_ _5"></span>A<span class="_ _7"></span>TE<span class="_ _6"> </span>dri<span class="_ _4"></span>ves<span class="_ _6"> </span>(SSDs)<span class="_ _b"> </span>based<span class="_ _6"> </span>on<span class="_ _6"> </span><span class="fs6 lsc">NA<span class="_ _8"></span>N<span class="_ _8"></span>D<span class="_ _6"> </span></span>&#64258;ash<span class="_ _b"> </span>mem-</div><div class="t m0 x9 h6 y21 ff1 fs5 fc0 sc0 lsb ws0">ory<span class="_ _6"> </span>ha<span class="_ _4"></span>ve<span class="_ _6"> </span>become<span class="_ _b"> </span>popular<span class="_ _6"> </span>in<span class="_ _b"> </span>consumer<span class="_ _6"> </span>electronic<span class="_ _b"> </span>devices</div><div class="t m0 x1 h6 y22 ff1 fs5 fc0 sc0 lsb ws0">such<span class="_ _e"> </span>as<span class="_ _e"> </span>smart<span class="_ _e"> </span>phones,<span class="_ _e"> </span>tablet,<span class="_ _e"> </span>and<span class="_ _e"> </span>desktop<span class="_ _e"> </span>systems<span class="_ _6"> </span>[<span class="fc1 ls1">1</span><span class="lsd">],<span class="_ _b"> </span>[<span class="fc1 ls1">2</span><span class="lse">].</span></span></div><div class="t m0 x1 h6 y23 ff1 fs5 fc0 sc0 lsb ws0">It<span class="_ _c"> </span>is<span class="_ _c"> </span>highly<span class="_ _c"> </span>desirable<span class="_ _c"> </span>to<span class="_ _c"> </span>reduce<span class="_ _c"> </span>the<span class="_ _c"> </span>amount<span class="_ _12"> </span>of<span class="_ _c"> </span>data<span class="_ _c"> </span>in<span class="_ _c"> </span>SSDs</div><div class="t m0 x1 h6 y24 ff1 fs5 fc0 sc0 lsf ws0">and<span class="_ _d"> </span>the<span class="_ _e"> </span>read/write<span class="_ _e"> </span>data<span class="_ _e"> </span>transmission<span class="_ _d"> </span>time<span class="_ _e"> </span>to/from<span class="_ _e"> </span>SSDs<span class="_ _d"> </span>as</div><div class="t m0 x1 h6 y25 ff1 fs5 fc0 sc0 lsb ws0">&#64258;ash<span class="_ _f"> </span>memory<span class="_ _f"> </span>has<span class="_ _10"> </span>a<span class="_ _f"> </span>&#64257;nite<span class="_ _10"> </span>number<span class="_ _f"> </span>of<span class="_ _f"> </span>program-erase<span class="_ _10"> </span>(P/E)</div><div class="t m0 x1 h6 y26 ff1 fs5 fc0 sc0 lsf ws0">cycles<span class="_ _e"> </span>thus<span class="_ _e"> </span>limited<span class="_ _e"> </span>lifetime<span class="_ _6"> </span>[<span class="fc1 ls1">3</span><span class="lsb">].<span class="_ _d"> </span>For<span class="_ _e"> </span>example,<span class="_ _e"> </span>older<span class="_ _e"> </span>single-</span></div><div class="t m0 x1 h6 y27 ff1 fs5 fc0 sc0 lsb ws0">le<span class="_ _2"></span>vel<span class="_ _c"> </span>cell<span class="_ _c"> </span>(SLC)</div><div class="t m0 xa h6 y28 ff1 fs6 fc0 sc0 ls10 ws0">N<span class="_ _2"></span>AND<span class="fs5 lsf">-&#64258;ash<span class="_ _6"> </span>memory<span class="_ _12"> </span>was<span class="_ _c"> </span>able<span class="_ _c"> </span>to<span class="_ _12"> </span>withstand</span></div><div class="t m0 x1 h6 y29 ff1 fs5 fc0 sc0 lsf ws0">150<span class="_ _b"> </span>000<span class="_ _6"> </span>P/E<span class="_ _b"> </span>cycles,<span class="_ _b"> </span>while<span class="_ _b"> </span>multilev<span class="_ _2"></span>el<span class="_ _b"> </span>cell<span class="_ _6"> </span>(MLC)<span class="_ _b"> </span><span class="fs6 lsc">NA<span class="_ _8"></span>N<span class="_ _8"></span>D</span><span class="ls1">-&#64258;ash</span></div><div class="t m0 x1 h6 y2a ff1 fs5 fc0 sc0 lsb ws0">memory<span class="_ _a"> </span>using<span class="_ _9"> </span>15&#8211;19<span class="_ _6"> </span>nm<span class="_ _a"> </span>process<span class="_ _a"> </span>technologies<span class="_ _9"> </span>wears<span class="_ _9"> </span>out<span class="_ _a"> </span>after</div><div class="t m0 x1 h6 y2b ff1 fs5 fc0 sc0 ls11 ws0">only<span class="_ _6"> </span>3000<span class="_ _c"> </span>P/E<span class="_ _c"> </span>cycles<span class="_ _b"> </span>[<span class="fc1 ls1">2</span><span class="lse">],<span class="_ _b"> </span>[<span class="fc1 ls1">4</span><span class="lsf">].<span class="_ _c"> </span>Furthermore,<span class="_ _c"> </span>the<span class="_ _c"> </span>performance</span></span></div><div class="t m0 xb h8 y2c ff1 fs6 fc0 sc0 ls12 ws0">Manuscript<span class="_ _f"> </span>received<span class="_ _f"> </span>December<span class="_ _f"> </span>4,<span class="_ _f"> </span>2017;<span class="_ _f"> </span>revised<span class="_ _f"> </span>February<span class="_ _f"> </span>10,<span class="_ _10"> </span>2018;</div><div class="t m0 x1 h8 y2d ff1 fs6 fc0 sc0 ls1 ws0">accepted<span class="_ _1"> </span>February<span class="_ _a"> </span>15,<span class="_ _1"> </span>2018.<span class="_ _a"> </span>Date<span class="_ _1"> </span>of<span class="_ _a"> </span>publication<span class="_ _1"> </span>March<span class="_ _a"> </span>2,<span class="_ _1"> </span>2018;<span class="_ _a"> </span>date<span class="_ _1"> </span>of<span class="_ _a"> </span>cur-</div><div class="t m0 x1 h8 y2e ff1 fs6 fc0 sc0 ls12 ws0">rent<span class="_ _a"> </span>version<span class="_ _9"> </span>March<span class="_ _9"> </span>29,<span class="_ _9"> </span>2018.<span class="_ _a"> </span>This<span class="_ _9"> </span>work<span class="_ _9"> </span>was<span class="_ _a"> </span>supported<span class="_ _9"> </span>by<span class="_ _9"> </span>the<span class="_ _9"> </span>Fundamental</div><div class="t m0 x1 h8 y2f ff1 fs6 fc0 sc0 ls12 ws0">Research<span class="_ _a"> </span>Funds<span class="_ _9"> </span>for<span class="_ _a"> </span>the<span class="_ _9"> </span>Central<span class="_ _a"> </span>Universities<span class="_ _a"> </span>China<span class="_ _a"> </span>under<span class="_ _9"> </span>Grant<span class="_ _a"> </span>NS2017024.</div><div class="t m0 x1 h9 y30 ff2 fs6 fc0 sc0 ls12 ws0">(Corr<span class="_ _2"></span>esponding<span class="_ _a"> </span>author:<span class="_ _a"> </span>W<span class="_ _5"></span>eiqiang<span class="_ _a"> </span>Liu.)</div><div class="t m0 xb h8 y31 ff1 fs6 fc0 sc0 ls1 ws0">W<span class="_ _5"></span>.<span class="_ _10"> </span>Liu,<span class="_ _f"> </span>F<span class="_ _4"></span>.<span class="_ _f"> </span>Mei,<span class="_ _10"> </span>and<span class="_ _f"> </span>C.<span class="_ _10"> </span>W<span class="_ _4"></span>ang<span class="_ _f"> </span>are<span class="_ _10"> </span>with<span class="_ _f"> </span>the<span class="_ _10"> </span>College<span class="_ _f"> </span>of<span class="_ _f"> </span>Electronic</div><div class="t m0 x1 h8 y32 ff1 fs6 fc0 sc0 ls1 ws0">and<span class="_ _13"> </span>Information<span class="_ _13"> </span>Engineering,<span class="_ _13"> </span>Nanjing<span class="_ _13"> </span>Univ<span class="_ _2"></span>ersity<span class="_ _13"> </span>of<span class="_ _13"> </span>Aeronautics<span class="_ _13"> </span>and</div><div class="t m0 x1 h8 y33 ff1 fs6 fc0 sc0 ls1 ws0">Astronautics,<span class="_ _f"> </span>Nanjing<span class="_ _10"> </span>211106,<span class="_ _10"> </span>China<span class="_ _f"> </span>(e-mail:<span class="_ _10"> </span>liuweiqiang@nuaa.edu.cn;</div><div class="t m0 x1 h8 y34 ff1 fs6 fc0 sc0 ls1 ws0">meifaqiang@nuaa.edu.cn;<span class="_ _a"> </span>chwang@nuaa.edu.cn).</div><div class="t m0 xb h8 y35 ff1 fs6 fc0 sc0 ls1 ws0">M.<span class="_"> </span>O&#8217;Neill<span class="_"> </span>is<span class="_"> </span>with<span class="_"> </span>the<span class="_"> </span>Center<span class="_"> </span>for<span class="_"> </span>Secure<span class="_"> </span>Information<span class="_"> </span>T<span class="_ _5"></span>echnologies,<span class="_"> </span>Queen&#8217;<span class="_ _4"></span>s</div><div class="t m0 x1 h8 y36 ff1 fs6 fc0 sc0 ls13 ws0">Univ<span class="_ _2"></span>ersity<span class="_ _1"> </span>Belfast,<span class="_ _1"> </span>Belfast<span class="_ _1"> </span>BT3<span class="_ _1"> </span>9DT<span class="_ _5"></span>,<span class="_ _1"> </span>U.K.<span class="_ _1"> </span>(e-mail:<span class="_ _a"> </span>m.oneill@ecit.qub<span class="_ _4"></span>.ac.uk).</div><div class="t m0 xb h8 y37 ff1 fs6 fc0 sc0 ls13 ws0">E.<span class="_ _6"> </span>E.<span class="_ _c"> </span>Swartzlander<span class="_ _6"> </span>is<span class="_ _c"> </span>with<span class="_ _6"> </span>the<span class="_ _c"> </span>Department<span class="_ _6"> </span>of<span class="_ _c"> </span>Electrical<span class="_ _6"> </span>and<span class="_ _c"> </span>Computer</div><div class="t m0 x1 h8 y38 ff1 fs6 fc0 sc0 ls1 ws0">Engineering,<span class="_ _a"> </span>Univ<span class="_ _2"></span>ersity<span class="_ _a"> </span>of<span class="_ _a"> </span>T<span class="_ _4"></span>exas<span class="_ _1"> </span>at<span class="_ _a"> </span>Austin,<span class="_ _a"> </span>Austin,<span class="_ _9"> </span>TX<span class="_ _a"> </span>78712<span class="_ _a"> </span>USA<span class="_ _a"> </span>(e-mail:</div><div class="t m0 x1 h8 y39 ff1 fs6 fc0 sc0 ls13 ws0">eswartzla@aol.com).</div><div class="t m0 xb h8 y3a ff1 fs6 fc0 sc0 ls1 ws0">Digital<span class="_ _a"> </span>Object<span class="_ _a"> </span>Identi&#64257;er<span class="_ _a"> </span>10.1109/TCE.2018.2810480</div><div class="t m0 xc h8 y3b ff1 fs6 fc0 sc0 ls13 ws0">Fig.<span class="_ _a"> </span>1.<span class="_ _3"> </span>T<span class="_ _4"></span>ypical<span class="_ _a"> </span>SSD<span class="_ _a"> </span>architecture<span class="_ _a"> </span>with<span class="_ _a"> </span>data<span class="_ _a"> </span>compression<span class="_ _a"> </span>acceleration.</div><div class="t m0 xc h6 y3c ff1 fs5 fc0 sc0 lsb ws0">of<span class="_ _12"> </span>MLC<span class="_ _c"> </span>&#64258;ash<span class="_ _12"> </span>memory<span class="_ _12"> </span>is<span class="_ _12"> </span>also<span class="_ _12"> </span>much<span class="_ _12"> </span>slower<span class="_ _c"> </span>than<span class="_ _12"> </span>that<span class="_ _12"> </span>of<span class="_ _c"> </span>its</div><div class="t m0 xc h6 y3d ff1 fs5 fc0 sc0 lsb ws0">SLC<span class="_ _b"> </span>counterpart.<span class="_ _b"> </span>Also,<span class="_ _b"> </span>more<span class="_ _b"> </span>advanced<span class="_ _9"> </span>triple-lev<span class="_ _2"></span>el<span class="_ _b"> </span>cell</div><div class="t m0 xd h8 y3e ff1 fs6 fc0 sc0 lsc ws0">NA<span class="_ _8"></span>ND</div><div class="t m0 xc h6 y3f ff1 fs5 fc0 sc0 ls11 ws0">&#64258;ash<span class="_ _12"> </span>memory<span class="_ _12"> </span>has<span class="_ _12"> </span>an<span class="_ _12"> </span>ev<span class="_ _2"></span>en<span class="_ _12"> </span>lower<span class="_ _c"> </span>number<span class="_ _12"> </span>of<span class="_ _12"> </span>P/E<span class="_ _d"> </span>cycles<span class="_ _b"> </span>[<span class="fc1 ls1">5</span><span class="ls14">].</span></div><div class="t m0 xc h6 y40 ff1 fs5 fc0 sc0 lsf ws0">This<span class="_ _12"> </span>problem<span class="_ _12"> </span>is<span class="_ _12"> </span>expected<span class="_ _12"> </span>to<span class="_ _d"> </span>worsen<span class="_ _12"> </span>with<span class="_ _12"> </span>further<span class="_ _12"> </span>scaling<span class="_ _12"> </span>of</div><div class="t m0 xc h6 y41 ff1 fs5 fc0 sc0 lsf ws0">the<span class="_ _9"> </span>semiconductor<span class="_ _b"> </span>process.<span class="_ _b"> </span>Therefore,<span class="_ _9"> </span>to<span class="_ _b"> </span>increase<span class="_ _9"> </span>the<span class="_ _b"> </span>lifetime</div><div class="t m0 xc h6 y42 ff1 fs5 fc0 sc0 lsb ws0">and<span class="_ _b"> </span>also<span class="_ _b"> </span>the<span class="_ _b"> </span>performance<span class="_ _b"> </span>of<span class="_ _b"> </span>&#64258;ash-based<span class="_ _b"> </span>SSDs,<span class="_ _b"> </span>the<span class="_ _b"> </span>amount<span class="_ _b"> </span>of</div><div class="t m0 xc h6 y43 ff1 fs5 fc0 sc0 lsb ws0">data<span class="_ _c"> </span>written<span class="_ _12"> </span>to<span class="_ _12"> </span>and<span class="_ _c"> </span>read<span class="_ _12"> </span>from<span class="_ _12"> </span>the<span class="_ _c"> </span>SSDs<span class="_ _12"> </span>should<span class="_ _12"> </span>be<span class="_ _12"> </span>reduced,</div><div class="t m0 xc h6 y44 ff1 fs5 fc0 sc0 lsb ws0">which<span class="_ _b"> </span>can<span class="_ _b"> </span>be<span class="_ _b"> </span>achiev<span class="_ _2"></span>ed<span class="_ _b"> </span>using<span class="_ _b"> </span>data<span class="_ _b"> </span>compression.<span class="_ _6"> </span>Another<span class="_ _9"> </span>ben-</div><div class="t m0 xc h6 y45 ff1 fs5 fc0 sc0 lsb ws0">e&#64257;t<span class="_ _6"> </span>of<span class="_ _c"> </span>using<span class="_ _c"> </span>lossless<span class="_ _6"> </span>data<span class="_ _c"> </span>compression<span class="_ _6"> </span>in<span class="_ _c"> </span>SSDs<span class="_ _c"> </span>is<span class="_ _6"> </span>to<span class="_ _c"> </span>reduce</div><div class="t m0 xc h6 y46 ff1 fs5 fc0 sc0 lsf ws0">the<span class="_ _b"> </span>I/O<span class="_ _6"> </span>latenc<span class="_ _4"></span>y<span class="_ _4"></span>.</div><div class="t m0 xe h6 y47 ff1 fs5 fc0 sc0 lsb ws0">Data<span class="_ _13"> </span>compression<span class="_ _13"> </span>for<span class="_ _13"> </span>SSDs<span class="_ _13"> </span>has<span class="_ _13"> </span>been<span class="_ _13"> </span>widely<span class="_ _13"> </span>adopted.</div><div class="t m0 xc h6 y48 ff1 fs5 fc0 sc0 lsf ws0">Data<span class="_ _b"> </span>compression<span class="_ _6"> </span>can<span class="_ _b"> </span>be<span class="_ _6"> </span>implemented<span class="_ _b"> </span>in<span class="_ _6"> </span>three<span class="_ _b"> </span>layers:<span class="_ _b"> </span>1)<span class="_ _6"> </span>the</div><div class="t m0 xc h6 y49 ff1 fs5 fc0 sc0 lsf ws0">application;<span class="_ _b"> </span>2)<span class="_ _b"> </span>the<span class="_ _b"> </span>&#64257;le<span class="_ _b"> </span>system;<span class="_ _b"> </span>or<span class="_ _6"> </span>3)<span class="_ _9"> </span>the<span class="_ _b"> </span>&#64257;rmware<span class="_ _b"> </span>of<span class="_ _b"> </span>the<span class="_ _b"> </span>stor-</div><div class="t m0 xc h6 y4a ff1 fs5 fc0 sc0 lsb ws0">age<span class="_ _b"> </span>device.<span class="_ _b"> </span>Most<span class="_ _b"> </span>data<span class="_ _b"> </span>compression<span class="_ _6"> </span>algorithms<span class="_ _9"> </span>are<span class="_ _6"> </span>adopted<span class="_ _9"> </span>in</div><div class="t m0 xc h6 y4b ff1 fs5 fc0 sc0 lsf ws0">the<span class="_ _a"> </span>application<span class="_ _a"> </span>layer<span class="_ _9"> </span>and<span class="_ _a"> </span>the<span class="_ _9"> </span>&#64257;le<span class="_ _a"> </span>system<span class="_ _9"> </span>using<span class="_ _a"> </span>software<span class="_ _a"> </span>imple-</div><div class="t m0 xc h6 y4c ff1 fs5 fc0 sc0 lsf ws0">mentation.<span class="_ _b"> </span>Software-based<span class="_ _6"> </span>data<span class="_ _9"> </span>compression<span class="_ _6"> </span>can<span class="_ _b"> </span>be<span class="_ _6"> </span>useful<span class="_ _b"> </span>in</div><div class="t m0 xc h6 y4d ff1 fs5 fc0 sc0 lsf ws0">improving<span class="_ _b"> </span>the<span class="_ _6"> </span>lifetime<span class="_ _b"> </span>of<span class="_ _6"> </span>SSDs.<span class="_ _b"> </span>Howe<span class="_ _4"></span>ver<span class="_ _4"></span>,<span class="_ _6"> </span>the<span class="_ _b"> </span>overall<span class="_ _b"> </span>perfor-</div><div class="t m0 xc h6 y4e ff1 fs5 fc0 sc0 lsb ws0">mance<span class="_ _b"> </span>of<span class="_ _b"> </span>SSDs<span class="_ _b"> </span>is<span class="_ _b"> </span>reduced<span class="_ _6"> </span>signi&#64257;cantly<span class="_ _9"> </span>due<span class="_ _b"> </span>to<span class="_ _b"> </span>the<span class="_ _6"> </span>slo<span class="_ _4"></span>w<span class="_ _b"> </span>com-</div><div class="t m0 xc h6 y4f ff1 fs5 fc0 sc0 lsb ws0">pression<span class="_ _6"> </span>and<span class="_ _6"> </span>decompression<span class="_ _c"> </span>speed.<span class="_ _6"> </span>A<span class="_ _c"> </span>recent<span class="_ _6"> </span>study<span class="_ _6"> </span>[<span class="fc1 ls1">6</span>]<span class="_ _6"> </span>based</div><div class="t m0 xc h6 y50 ff1 fs5 fc0 sc0 lsb ws0">on<span class="_ _c"> </span>realistic<span class="_ _12"> </span>data<span class="_ _12"> </span>and<span class="_ _12"> </span>systems<span class="_ _12"> </span>show<span class="_ _c"> </span>that<span class="_ _12"> </span>applying<span class="_ _12"> </span>data<span class="_ _c"> </span>com-</div><div class="t m0 xc h6 y51 ff1 fs5 fc0 sc0 lsf ws0">pression<span class="_"> </span>in<span class="_ _a"> </span>the<span class="_"> </span>&#64257;rmware<span class="_ _a"> </span>of<span class="_"> </span>the<span class="_ _a"> </span>SSDs<span class="_"> </span>using<span class="_ _a"> </span>a<span class="_"> </span>data<span class="_ _a"> </span>compression</div><div class="t m0 xc h6 y52 ff1 fs5 fc0 sc0 lsb ws0">hardware<span class="_ _1"> </span>accelerator<span class="_ _11"> </span>is<span class="_"> </span>the<span class="_"> </span>best<span class="_ _11"> </span>approach.<span class="_"> </span>A<span class="_"> </span>typical<span class="_ _14"> </span>SSD<span class="_"> </span>archi-</div><div class="t m0 xc h6 y53 ff1 fs5 fc0 sc0 lsf ws0">tecture<span class="_ _9"> </span>with<span class="_ _9"> </span>data<span class="_ _9"> </span>compression<span class="_ _9"> </span>acceleration<span class="_ _9"> </span>is<span class="_ _b"> </span>sho<span class="_ _2"></span>wn<span class="_ _9"> </span>in<span class="_ _9"> </span>Fig.<span class="_ _9"> </span><span class="fc1 ls1">1<span class="fc0">.</span></span></div><div class="t m0 xe h6 y54 ff1 fs5 fc0 sc0 lsb ws0">Although<span class="_"> </span>hardware-based<span class="_ _11"> </span>compression<span class="_"> </span>is<span class="_"> </span>required<span class="_ _14"> </span>for</div><div class="t m0 xd h8 y55 ff1 fs6 fc0 sc0 lsa ws0">N<span class="_ _2"></span>AND</div><div class="t m0 xc h6 y56 ff1 fs5 fc0 sc0 lsb ws0">&#64258;ash<span class="_ _12"> </span>memory<span class="_ _12"> </span>and<span class="_ _12"> </span>SSDs,<span class="_ _12"> </span>little<span class="_ _12"> </span>research<span class="_ _12"> </span>has<span class="_ _12"> </span>been<span class="_ _12"> </span>conducted</div><div class="t m0 xc h6 y57 ff1 fs5 fc0 sc0 lsb ws0">on<span class="_ _f"> </span>how<span class="_ _f"> </span>to<span class="_ _f"> </span>design<span class="_ _10"> </span>a<span class="_ _10"> </span>high<span class="_ _f"> </span>performance<span class="_ _10"> </span>hardware<span class="_ _f"> </span>compres-</div><div class="t m0 xc h6 y58 ff1 fs5 fc0 sc0 lsb ws0">sion<span class="_ _c"> </span>accelerator<span class="_ _6"> </span>[<span class="fc1 ls1">7</span><span class="lsf">]&#8211;[<span class="fc1 ls15">13</span><span class="lsd">].<span class="_ _6"> </span>In<span class="_ _6"> </span>[<span class="fc1 ls1">6</span></span></span>],<span class="_ _6"> </span>it<span class="_ _12"> </span>was<span class="_ _c"> </span>found<span class="_ _c"> </span>that<span class="_ _12"> </span>for<span class="_ _c"> </span>high-</div><div class="t m0 xc h6 y59 ff1 fs5 fc0 sc0 lsb ws0">end<span class="_ _f"> </span>SSDs<span class="_ _f"> </span>with<span class="_ _f"> </span>transaction<span class="_ _f"> </span>rates<span class="_ _f"> </span>of<span class="_ _f"> </span>up<span class="_ _f"> </span>to<span class="_ _f"> </span>3K<span class="_ _f"> </span>per<span class="_ _f"> </span>second,</div><div class="t m0 xc h6 y5a ff1 fs5 fc0 sc0 lsf ws0">compression/decompression<span class="_ _10"> </span>rates<span class="_ _10"> </span>of<span class="_ _10"> </span>above<span class="_ _10"> </span>200<span class="_ _10"> </span>Mb/s<span class="_ _10"> </span>(i.e.,</div><div class="t m0 xc h6 y5b ff1 fs5 fc0 sc0 lsf ws0">1.6<span class="_ _e"> </span>Gb/s)<span class="_ _e"> </span>are<span class="_ _f"> </span>required.<span class="_ _e"> </span>Howe<span class="_ _4"></span>ver<span class="_ _4"></span>,<span class="_ _f"> </span>existing<span class="_ _d"> </span>designs<span class="_ _f"> </span>are<span class="_ _e"> </span>lim-</div><div class="t m0 xc h6 y5c ff1 fs5 fc0 sc0 lsf ws0">ited<span class="_ _6"> </span>in<span class="_ _b"> </span>performance<span class="_ _6"> </span>with<span class="_ _6"> </span>compression<span class="_ _b"> </span>speeds<span class="_ _6"> </span>in<span class="_ _6"> </span>the<span class="_ _b"> </span>range<span class="_ _6"> </span>of</div><div class="t m0 xc h6 y5d ff1 fs5 fc0 sc0 lsb ws0">0.567&#8211;1.6<span class="_ _6"> </span>Gb/s<span class="_ _b"> </span>[<span class="fc1 ls1">7<span class="fc0">]&#8211;[</span><span class="ls15">13<span class="fc0">],<span class="_ _6"> </span>which<span class="_ _6"> </span>cannot<span class="_ _6"> </span>meet<span class="_ _6"> </span>the<span class="_ _6"> </span>requirement</span></span></span></div><div class="t m0 xc h6 y5e ff1 fs5 fc0 sc0 lsb ws0">of<span class="_ _b"> </span>high-end<span class="_ _6"> </span>SSDs.</div><div class="t m0 xe h6 y5f ff1 fs5 fc0 sc0 lsb ws0">In<span class="_ _a"> </span>this<span class="_ _9"> </span>paper,<span class="_ _a"> </span>the<span class="_ _9"> </span>design<span class="_ _a"> </span>of<span class="_ _9"> </span>a<span class="_ _9"> </span>hardware<span class="_ _9"> </span>accelerator<span class="_ _9"> </span>based<span class="_ _a"> </span>on</div><div class="t m0 xc h6 y60 ff1 fs5 fc0 sc0 lsf ws0">the<span class="_"> </span>latest<span class="_"> </span>lossless<span class="_"> </span>data<span class="_"> </span>compression<span class="_"> </span>algorithm,<span class="_"> </span>i.e.,<span class="_"> </span>Lempel-Zi<span class="_ _4"></span>v</div><div class="t m0 xf h8 y61 ff1 fs6 fc0 sc0 ls16 ws0">1558-4127</div><div class="t m0 x10 h8 y62 ff1 fs6 fc0 sc0 ls1 ws0">c</div><div class="t m0 x11 h8 y61 ff5 fs6 fc0 sc0 ls1 ws0">&#58882;<span class="_"> </span><span class="ff1">2018<span class="_ _a"> </span>IEEE.<span class="_ _a"> </span>Personal<span class="_ _a"> </span>use<span class="_ _a"> </span>is<span class="_ _a"> </span>permitted,<span class="_ _9"> </span>but<span class="_ _1"> </span>republication/redistribution<span class="_ _a"> </span>requires<span class="_ _a"> </span>IEEE<span class="_ _a"> </span>permission.</span></div><div class="t m0 x12 h8 y63 ff1 fs6 fc0 sc0 ls1 ws0">See<span class="_ _a"> </span>http://www<span class="_ _4"></span>.ieee.org/publications_standards/publications/rights/index.html<span class="_ _1"> </span>for<span class="_ _a"> </span>more<span class="_ _a"> </span>information.</div><a class="l" rel='nofollow' onclick='return false;'><div class="d m1"></div></a><a class="l" rel='nofollow' onclick='return false;'><div class="d m1"></div></a><a class="l" rel='nofollow' onclick='return false;'><div class="d m1"></div></a><a class="l" rel='nofollow' onclick='return false;'><div class="d m1"></div></a><a class="l" rel='nofollow' onclick='return false;'><div class="d m1"></div></a><a class="l" rel='nofollow' onclick='return false;'><div class="d m1"></div></a><a class="l" rel='nofollow' onclick='return false;'><div class="d m1"></div></a><a class="l" rel='nofollow' onclick='return false;'><div class="d m1"></div></a><a class="l" rel='nofollow' onclick='return false;'><div class="d m1"></div></a><a class="l" rel='nofollow' onclick='return false;'><div class="d m1"></div></a><a class="l" rel='nofollow' onclick='return false;'><div class="d m1"></div></a><a class="l" rel='nofollow' onclick='return false;'><div class="d m1"></div></a><a class="l" rel='nofollow' onclick='return false;'><div class="d m1"></div></a></div><div class="pi" data-data='{"ctm":[1.568627,0.000000,0.000000,1.568627,0.000000,0.000000]}'></div></div> </body> </html>
评论
    相关推荐
    • lz4net-master.zip
      对应C#版本的LZ4 压缩工具,,,,,,,
    • lz4.rar
      LZ4 Kernel Interface for Linux v2.13.6.
    • lz4.zip
      1、独立lz4压缩移植,零修改,只添加用户自定义文件lz4_user.c和lz4_user.h; 2、默认用的栈资源(速度快),可选用堆。# define LZ4_HEAPMODE 0; 3、默认用到栈大小为14k,若资源有限可调小。# define LZ4_...
    • lz4hc.rar
      16 General propose registers, 32 bit on s 390 64 bit on z Architecture, r0-r15 or gpr0-gpr15 used for arithmetic & addressing.
    • lz4b
      lz4b:lz4 erlang绑定 框架API 帧压缩 lz4b_frame : compress ( binary ()) -> { ok , binary ()} | { error , atom ()}. 帧解压缩 lz4b_frame : decompress ( binary ()) -> { ok , binary ()} | { error , atom ()...
    • LZ4Delphi:LZ4德尔福
      LZ4德尔福 LZ4Delphi,与LZ4 r 127兼容 包括 1) LZ4 压缩算法的 Delphi 绑定,带有 VS 2012 .obj 文件。 32 位和 64 位 2)作为替代:原始LZ4 C源的Delphi转换。 工作速度比上面的 1) 慢一点,但仍然很快。 3)...
    • lz4:http的自动镜像
      LZ4 - 极快的压缩 LZ4 是无损压缩算法,提供每核 400 MB/s 的压缩速度,可通过多核 CPU 进行扩展。 它还具有极快的解码器,每个内核的速度达数 GB/s,通常达到多核系统上的 RAM 速度限制。 还提供了一种称为 LZ4_HC ...
    • LZ4极速压缩
      关于LZ4压缩算法可到百度搜索以了解详情。 此程序用VS2012编译生成,如若不能运行请到此链接下载对应版本的运行时库安装即可: http://www.microsoft.com/zh-CN/download/details.aspx?id=30679 win8以上系统应该...
    • lz4.zip
      his specification is intended for developers willing to produce LZ4-compatible compressed data blocks using any programming language. LZ4 is an LZ77-type compressor with a fixed, byte-oriented ...
    • lzw_soft.zip
      lzw压缩解压算法源码