.. başlıq: TCP Congestion Control Müqayisəsi
.. şlak: tcp-congestion-control-müqayisə
.. tarix: 2026-03-27 12:00:00 UTC
.. teqlər: şəbəkə, tcp, tıxaclara nəzarət, bbr, kub, performans, kalkulyator, alət
.. kateqoriya: Kalkulyatorlar
.. keçid:
.. təsviri: CUBIC, BBR və Reno TCP tıxanmaya nəzarət alqoritmlərini dəyişən itki dərəcəsi və RTT altında müqayisə edin. Məhsuldarlıq modelləri və iş yükü tövsiyə matrisi.
.. növü: mətn
<style>
/* ── CANONICAL DESIGN SYSTEM ──────────────────────────────────────────────── */
.calc-container { max-width: 900px; margin: 20px auto; padding: 20px; }
.calc-header { margin-bottom: 30px; padding: 25px; background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); border-radius: 12px; box-shadow: 0 4px 6px rgba(0,0,0,0.1); color: white; }
.calc-header h2 { margin-top: 0; color: white; font-size: 28px; margin-bottom: 10px; }
.calc-header p  { margin: 10px 0; opacity: 0.95; font-size: 15px; }
.input-grid { display: grid; grid-template-columns: repeat(auto-fit, minmax(220px, 1fr)); gap: 15px; margin-top: 20px; }
.input-group { display: flex; flex-direction: column; }
.input-group label { margin-bottom: 8px; font-weight: 600; color: rgba(255,255,255,0.95); font-size: 14px; }
.input-group input, .input-group select, .input-group textarea { padding: 14px; font-size: 14px; border: 2px solid rgba(255,255,255,0.3); border-radius: 6px; background: rgba(255,255,255,0.95); transition: all 0.3s ease; box-sizing: border-box; font-family: inherit; }
.input-group input:focus, .input-group select:focus, .input-group textarea:focus { outline: none; border-color: #4CAF50; background: white; box-shadow: 0 0 0 3px rgba(76,175,80,0.1); }
.button-container { margin-top: 20px; display: flex; justify-content: center; gap: 10px; flex-wrap: wrap; }
.calc-button { padding: 14px 35px; font-size: 16px; background: #4CAF50; color: white; border: none; border-radius: 6px; cursor: pointer; font-weight: bold; transition: all 0.3s ease; box-shadow: 0 2px 4px rgba(0,0,0,0.2); }
.calc-button:hover  { background: #45a049; transform: translateY(-1px); box-shadow: 0 4px 8px rgba(0,0,0,0.3); }
.calc-button:active { background: #3d8b40; transform: translateY(0); }
.result-section { background: white; border: 1px solid #e0e0e0; border-radius: 8px; margin-top: 20px; overflow: hidden; box-shadow: 0 2px 4px rgba(0,0,0,0.05); transition: all 0.3s ease; }
.result-section:hover { box-shadow: 0 4px 8px rgba(0,0,0,0.1); }
.section-header { background: linear-gradient(135deg, #4CAF50 0%, #45a049 100%); color: white; padding: 15px 20px; font-size: 18px; font-weight: bold; }
.section-content { padding: 20px; background: #fafafa; }
.info-box { margin-top: 15px; padding: 15px; background: #e3f2fd; color: #1565c0; border-radius: 8px; border-left: 4px solid #2196F3; font-size: 14px; line-height: 1.6; }
table { width: 100%; border-collapse: collapse; }
th { background: linear-gradient(135deg, #4CAF50 0%, #45a049 100%); color: white; padding: 10px; text-align: left; }
td { padding: 8px 10px; border-bottom: 1px solid #eee; font-size: 13px; }
tr:hover td { background: #f8fff8; }
@media (max-width: 768px) { .input-grid, .algo-grid { grid-template-columns: 1fr; } .calc-header h2 { font-size: 22px; } .calc-container { padding: 10px; } }
/* Page-specific */
.algo-grid { display: grid; grid-template-columns: repeat(auto-fit, minmax(220px, 1fr)); gap: 16px; }
.algo-card { background: white; border-radius: 8px; border: 2px solid #4CAF50; padding: 16px; box-shadow: 0 1px 3px rgba(0,0,0,0.05); }
.algo-card h4 { margin-top: 0; color: #2d6a2d; font-size: 15px; }
.algo-label { font-size: 11px; font-weight: bold; color: #555; text-transform: uppercase; letter-spacing: 0.5px; margin-bottom: 2px; }
.algo-value { font-family: 'Courier New', monospace; font-size: 22px; font-weight: bold; color: #4CAF50; }
.bar-container { background: #e0e0e0; border-radius: 4px; height: 12px; overflow: hidden; margin-top: 6px; }
.bar-fill { background: #4CAF50; height: 100%; transition: width 0.4s ease; }
.metric { display: flex; justify-content: space-between; font-size: 13px; padding: 3px 0; border-bottom: 1px solid rgba(0,0,0,0.08); }
.metric:last-child { border: none; }
.metric-val { font-weight: bold; }
.bar-row { display: flex; align-items: center; gap: 8px; margin: 4px 0; font-size: 13px; }
.bar-label { width: 60px; font-weight: bold; }
.bar-outer { flex: 1; background: #eee; border-radius: 4px; height: 20px; }
.bar-val { width: 80px; text-align: right; font-size: 12px; }
.rec { display: inline-block; padding: 2px 8px; border-radius: 10px; font-size: 11px; font-weight: bold; }
.rec-bbr { background: #e8f5e9; color: #1b5e20; }
.rec-cubic { background: #e3f2fd; color: #0d47a1; }
.rec-reno { background: #fff3e0; color: #e65100; }
</style>
<div class="calc-container">
<div class="calc-header">
<h2>TCP Tıxacına Nəzarət Müqayisəsi</h2>
<p>CUBIC, BBR və Reno ötürmə qabiliyyətini dəyişən itki dərəcəsi və RTT altında müqayisə edin. İş yükü tövsiyə matrisi daxildir.</p>
<div class="input-grid">
<div class="input-group">
<label>RTT (ms)</label>
<input id="rttMs" max="2000" min="1" type="number" value="100"/>
</div>
<div class="input-group">
<label>Paket itkisi dərəcəsi (%)</label>
<input id="lossPct" max="20" min="0" step="0.01" type="number" value="0.1"/>
</div>
<div class="input-group">
<label>Bant genişliyi (Mbps)</label>
<input id="bwMbps" max="100000" min="0.1" step="any" type="number" value="100"/>
</div>
</div>
<div class="button-container">
<button class="calc-button" onclick="compare()">Müqayisə et</button>
</div>
</div>
<div id="compareResult" style="display:none"></div>
<div class="result-section">
<div class="section-header">İş Yükü Tövsiyə Matrisi</div>
<div class="section-content">
<table>
<tr><th>İş yükü</th><th>Tövsiyə olunur</th><th>çəkinin</th><th>Səbəb</th></tr>
<tr><td>Toplu WAN transferi (yüksək BDP)</td><td><span class="rec rec-bbr">BBR</span></td>
<td>Reno</td><td>BBR növbələri şişirtmədən BDP-ni doldurur; Reno yüksək RTT × BW məhsullarında dayanır</td></tr>
<tr><td>İnteraktiv SSH / Telnet</td><td><span class="rec rec-cubic">KUB</span></td>
<td>—</td><td>Aşağı BDP; bütün alqoritmlər oxşar; Linux-da CUBIC defolt yaxşıdır</td></tr>
<tr><td>Video axını (davamlı)</td><td><span class="rec rec-bbr">BBR</span></td>
<td>Reno</td><td>BBR itkiyə səbəb olan düşmələr olmadan sabit göndərmə sürətini saxlayır</td></tr>
<tr><td>Datacenter şərq-qərb (aşağı RTT)</td><td><span class="rec rec-cubic">KUB</span></td>
<td>BBR</td><td>Çox aşağı RTT; CUBIC sürətlə birləşir; BBR-nin yoxlaması lazımsız titrəmə əlavə edir</td></tr>
<tr><td>Yüksək itkili mobil əlaqə</td><td><span class="rec rec-bbr">BBR</span></td>
<td><span class="rec rec-reno">Reno</span></td><td>BBR zərərə əsaslanmayan BW təxminindən istifadə edir; simsiz düşmələri tıxac kimi səhv şərh etmir</td></tr>
<tr><td>Köhnə son nöqtələrlə ədalətlilik</td><td><span class="rec rec-cubic">KUB</span></td>
<td>BBR</td><td>BBR paylaşılan darboğazlarda CUBIC/Reno axınlarına ədalətsiz hakim ola bilər (BBRv1 məsələsi)</td></tr>
</table>
</div>
</div>
<div class="result-section">
<div class="section-header">Alqoritmə Baxış</div>
<div class="section-content">
<table>
<tr><th>Alqoritm</th><th>yanaşma</th><th>Tıxac siqnalı</th><th>Üçün ən yaxşısı</th></tr>
<tr><td>KUB</td><td>İtirdikdən sonra cwnd-in kub polinom artımı</td><td>Paket itkisi</td><td>Ümumi təyinatlı; 2.6.19-dan bəri Linux defolt</td></tr>
<tr><td>BBR</td><td>Model əsaslı, BW + RTT darboğazını təxmin edir</td><td>BW doyma + RTT artımı</td><td>Yüksək BDP WAN, itkili bağlantılar, video axını</td></tr>
<tr><td>Reno</td><td>AIMD — əlavə artım, multiplikativ azalma</td><td>Paket itkisi</td><td>Köhnə uyğunluq; yüksək BDP bağlantılarından çəkinin</td></tr>
</table>
</div>
</div>
</div>
<script>
// CUBIC closed-form throughput approximation:
// throughput ≈ (MSS/RTT) * sqrt(3/2) / sqrt(loss) * C^(1/4) * RTT^(-3/4)
// Simplified: use Mathis with CUBIC scaling factor ~1.2x Reno
// BBR: targets BDP = BW × RTT, throughput = min(BW, BDP/RTT) = BW when not bottlenecked
// Reno (Mathis): throughput = (MSS/RTT) * (1/sqrt(loss))

function mathisBps(mss, rttS, lossFrac) {
    return lossFrac > 0 ? (mss / rttS) / Math.sqrt(lossFrac) : Infinity;
}

function calcReno(rttMs, lossFrac, bwBps) {
    const mss = 1460;
    const rttS = rttMs / 1000;
    const bps = Math.min(mathisBps(mss, rttS, lossFrac), bwBps);
    return { bps, mbps: bps / 1e6, utilPct: Math.min(bps / bwBps * 100, 100) };
}

function calcCubic(rttMs, lossFrac, bwBps) {
    const mss = 1460;
    const rttS = rttMs / 1000;
    // CUBIC throughput (Ha et al. 2008):
    // T_cubic = sqrt(3β/(2-β)) × C^(-1/4) × MSS / (RTT × sqrt(loss))
    // β=0.7, C=0.4 → factor relative to simplified Mathis ≈ 1.60
    // sqrt(3×0.7/(2-0.7)) / 0.4^0.25 = sqrt(1.615) / 0.795 ≈ 1.271/0.795 ≈ 1.60
    const cubicFactor = Math.sqrt(3 * 0.7 / (2 - 0.7)) / Math.pow(0.4, 0.25); // ≈ 1.60
    const bps = lossFrac > 0
        ? Math.min(cubicFactor * mathisBps(mss, rttS, lossFrac), bwBps)
        : bwBps;
    return { bps, mbps: bps / 1e6, utilPct: Math.min(bps / bwBps * 100, 100) };
}

function calcBbr(rttMs, lossFrac, bwBps) {
    // BBR does not reduce cwnd on loss (unlike CUBIC/Reno).
    // However, retransmissions consume bandwidth, reducing goodput.
    // Empirical model (Hock et al. 2017): goodput ≈ BW × (1 - 1.2 × loss)
    // The 1.2× factor accounts for each lost packet requiring ~1.2 transmit attempts on average.
    // Collapse floor at 5% minimum to avoid negative values at extreme loss.
    const goodput = Math.max(1.0 - 1.2 * lossFrac, 0.05);
    const bps = bwBps * goodput;
    return { bps, mbps: bps / 1e6, utilPct: Math.min(bps / bwBps * 100, 100) };
}

function compare() {
    const rttMs = parseFloat(document.getElementById('rttMs').value);
    const lossPct = parseFloat(document.getElementById('lossPct').value);
    const bwMbps = parseFloat(document.getElementById('bwMbps').value);
    const div = document.getElementById('compareResult');
    div.style.display = 'block';

    if (isNaN(rttMs) || isNaN(lossPct) || isNaN(bwMbps) || rttMs <= 0 || bwMbps <= 0 || lossPct < 0) {
        div.innerHTML = '<strong style="color:red">Invalid input.</strong>';
        return;
    }

    const lossFrac = lossPct / 100;
    const bwBps = bwMbps * 1e6;

    const reno = calcReno(rttMs, lossFrac, bwBps);
    const cubic = calcCubic(rttMs, lossFrac, bwBps);
    const bbr = calcBbr(rttMs, lossFrac, bwBps);

    const maxMbps = Math.max(reno.mbps, cubic.mbps, bbr.mbps, 0.001);
    const fmtMbps = v => v >= 1000 ? `${(v/1000).toFixed(1)} Gbps` : `${v.toFixed(2)} Mbps`;

    const winner = bbr.mbps >= cubic.mbps && bbr.mbps >= reno.mbps ? 'BBR'
        : cubic.mbps >= bbr.mbps && cubic.mbps >= reno.mbps ? 'CUBIC' : 'Reno';

    div.innerHTML = `
    <div class="result-section">
        <div class="section-header">Algorithm Comparison</div>
        <div class="section-content">
            <div class="algo-grid">
        <div class="algo-card cubic-card">
            <h4>CUBIC</h4>
            <div class="metric"><span>Throughput</span><span class="metric-val">${fmtMbps(cubic.mbps)}</span></div>
            <div class="metric"><span>Utilization</span><span class="metric-val">${cubic.utilPct.toFixed(1)}%</span></div>
            <div class="metric"><span>Loss response</span><span class="metric-val">cwnd halve</span></div>
        </div>
        <div class="algo-card bbr-card">
            <h4>BBR</h4>
            <div class="metric"><span>Throughput</span><span class="metric-val">${fmtMbps(bbr.mbps)}</span></div>
            <div class="metric"><span>Utilization</span><span class="metric-val">${bbr.utilPct.toFixed(1)}%</span></div>
            <div class="metric"><span>Loss response</span><span class="metric-val">BW model</span></div>
        </div>
        <div class="algo-card reno-card">
            <h4>Reno</h4>
            <div class="metric"><span>Throughput</span><span class="metric-val">${fmtMbps(reno.mbps)}</span></div>
            <div class="metric"><span>Utilization</span><span class="metric-val">${reno.utilPct.toFixed(1)}%</span></div>
            <div class="metric"><span>Loss response</span><span class="metric-val">cwnd halve</span></div>
        </div>
    </div>
    <div class="bar-container">
        <div class="bar-row"><span class="bar-label" style="color:#1976d2">CUBIC</span>
            <div class="bar-outer"><div class="bar-fill" style="background:#1976d2;width:${(cubic.mbps/maxMbps*100).toFixed(0)}%"></div></div>
            <span class="bar-val">${fmtMbps(cubic.mbps)}</span></div>
        <div class="bar-row"><span class="bar-label" style="color:#2e7d32">BBR</span>
            <div class="bar-outer"><div class="bar-fill" style="background:#2e7d32;width:${(bbr.mbps/maxMbps*100).toFixed(0)}%"></div></div>
            <span class="bar-val">${fmtMbps(bbr.mbps)}</span></div>
        <div class="bar-row"><span class="bar-label" style="color:#c62828">Reno</span>
            <div class="bar-outer"><div class="bar-fill" style="background:#c62828;width:${(reno.mbps/maxMbps*100).toFixed(0)}%"></div></div>
            <span class="bar-val">${fmtMbps(reno.mbps)}</span></div>
            </div>
            <div class="info-box">
                <strong>Winner at this operating point: ${winner}</strong><br>
                CUBIC uses cubic-polynomial model (Ha et al. 2008, β=0.7, C=0.4, ~1.6× Reno). BBR uses bandwidth estimation — retransmit overhead reduces goodput linearly (~1.2× loss rate). Reno uses simplified Mathis equation. These are analytical approximations — actual behavior depends on queue depth, congestion algorithm version, and competing flows.
            </div>
        </div>
    </div>`;
}
</script>