Knowledge (XXG)

Blue (queue management algorithm)

Source 📝

155:(DDoS) attacks. A resilient stochastic fair Blue (RSFB) algorithm was proposed in 2009 against spoofing DDoS attacks. The basic idea behind RSFB is to record the responsive normal TCP flows and rescue their dropped packets. RSFB algorithm is effective in preserving the TCP throughput in the presence of spoofing DDoS attacks. 123:
Stochastic fair Blue (SFB) is a stochastically fair variant of Blue which hashes flows and maintains a different mark/drop probability for each hash value. Assuming no hash collisions, SFB is able to provide a fair share of buffer space for every flow. In the presence of hash collisions, SFB is
139:, which dramatically reduces its storage requirements when the number of flows is large. When a flow's drop/mark probability reaches 1, the flow has been shown to not react to congestion indications from the network. Such an inelastic flow is put in a " 232:
Proceedings IEEE INFOCOM 2001. Conference on Computer Communications. Twentieth Annual Joint Conference of the IEEE Computer and Communications Society (Cat. No.01CH37213)
120:, but treats all flows as a single aggregate. Therefore, a single aggressive flow can push packets out of the queue belonging to other, better behaved, flows. 293: 63:
overflows. Unlike RED, however, it requires little or no tuning to be performed by the network administrator. A Blue queue maintains a drop/mark probability
229:
Wu-Chang Feng; Dilip D. Kandlur; Debanjan Saha; Kang G. Shin (April 2001). "Stochastic fair blue: A queue management algorithm for enforcing fairness".
441: 306: 261: 117: 56: 36: 389: 152: 446: 60: 377: 128: 239: 52: 32: 20: 244: 104:
will slowly converge to a value that keeps the queue within its bounds with full link utilization.
151:
Many scheduling algorithms, including the fairness-aimed ones, are notably vulnerable to spoofing
323: 267: 200: 113: 302: 257: 168: 24: 249: 414: 335: 340: 435: 271: 179: 132: 28: 230: 140: 253: 393: 295:
RSFB: a Resilient Stochastic Fair Blue algorithm against spoofing DDoS attacks
136: 199:
Wu-chang Feng; Dilip D. Kandlur; Debanjan Saha; Kang G. Shin (April 1999).
353: 127:
Unlike other stochastically fair queuing disciplines, such as SFQ (
175: 55:(RED), Blue operates by randomly dropping or marking packet with 164: 112:
The main flaw of Blue, which it shares with most single-queue
40: 27:
developed by graduate student Wu-chang Feng for Professor
201:"BLUE: A New Class of Active Queue Management Algorithms" 292:
Changwang Zhang; Jianping Yin & Zhiping Cai (2009).
100:
If the mix of traffic on the interface does not change,
71:
as they enter the queue. Whenever the queue overflows,
8: 415:"Stochastic Fair Blue for the Linux kernel" 390:"SFB Linux kernel network scheduler module" 67:, and drops/marks packets with probability 378:Kernel Newbies - Linux 2.6.39 - Networking 116:, is that it does not distinguish between 243: 210:(CSE–TR–387–99). University of Michigan 191: 59:mark before the transmit buffer of the 331: 321: 163:An implementation of Blue is part of 7: 238:. Vol. 3. pp. 1520–1529. 82:, and whenever the queue is empty, 131:), SFB can be implemented using a 14: 208:Computer Science Technical Report 75:is increased by a small constant 442:Packets (information technology) 57:explicit congestion notification 37:Thomas J. Watson Research Center 147:Resilient stochastic fair Blue 1: 174:An implementation of SFB for 153:distributed denial-of-service 61:network interface controller 129:Stochastic Fairness Queuing 86:is decreased by a constant 463: 254:10.1109/INFCOM.2001.916648 124:only stochastically fair. 301:. pp. 1566–1567. 53:random early detection 33:University of Michigan 143:", and rate-limited. 21:scheduling discipline 413:Juliusz Chroboczek. 178:was included in the 108:Stochastic fair Blue 447:Network performance 182:in version 2.6.39. 114:queuing disciplines 35:and others at the 308:978-1-4244-4521-9 263:978-0-7803-7016-6 169:network scheduler 25:network scheduler 454: 426: 425: 423: 421: 410: 404: 403: 401: 400: 386: 380: 375: 369: 368: 366: 364: 349: 343: 339: 333: 329: 327: 319: 317: 315: 300: 289: 283: 282: 280: 278: 247: 237: 226: 220: 219: 217: 215: 205: 196: 462: 461: 457: 456: 455: 453: 452: 451: 432: 431: 430: 429: 419: 417: 412: 411: 407: 398: 396: 388: 387: 383: 376: 372: 362: 360: 352:Wu-chang Feng. 351: 350: 346: 330: 320: 313: 311: 309: 298: 291: 290: 286: 276: 274: 264: 235: 228: 227: 223: 213: 211: 203: 198: 197: 193: 188: 161: 159:Implementations 149: 110: 95: 91: 80: 49: 12: 11: 5: 460: 458: 450: 449: 444: 434: 433: 428: 427: 405: 381: 370: 344: 332:|journal= 307: 284: 262: 245:10.1.1.11.4235 221: 190: 189: 187: 184: 171:for BSD Unix. 160: 157: 148: 145: 135:rather than a 109: 106: 93: 89: 78: 48: 45: 13: 10: 9: 6: 4: 3: 2: 459: 448: 445: 443: 440: 439: 437: 416: 409: 406: 395: 391: 385: 382: 379: 374: 371: 359: 355: 348: 345: 342: 337: 325: 310: 304: 297: 296: 288: 285: 273: 269: 265: 259: 255: 251: 246: 241: 234: 233: 225: 222: 209: 202: 195: 192: 185: 183: 181: 177: 172: 170: 166: 158: 156: 154: 146: 144: 142: 138: 134: 130: 125: 121: 119: 118:traffic flows 115: 107: 105: 103: 98: 96: 85: 81: 74: 70: 66: 62: 58: 54: 46: 44: 42: 38: 34: 30: 26: 22: 18: 418:. Retrieved 408: 397:. Retrieved 384: 373: 361:. Retrieved 357: 347: 312:. Retrieved 294: 287: 275:. Retrieved 231: 224: 212:. Retrieved 207: 194: 180:Linux kernel 173: 162: 150: 133:bloom filter 126: 122: 111: 101: 99: 92: < p 87: 83: 76: 72: 68: 64: 50: 29:Kang G. Shin 16: 15: 141:penalty box 47:Functioning 436:Categories 399:2013-09-07 394:kernel.org 186:References 137:hash table 334:ignored ( 324:cite book 240:CiteSeerX 43:in 1999. 358:Web page 341:Abstract 23:for the 420:June 8, 363:June 8, 314:June 8, 277:June 8, 272:5902623 214:June 8, 31:at the 354:"Blue" 305:  270:  260:  242:  167:, the 299:(PDF) 268:S2CID 236:(PDF) 204:(PDF) 176:Linux 51:Like 19:is a 422:2013 365:2013 336:help 316:2013 303:ISBN 279:2013 258:ISBN 216:2013 165:ALTQ 17:Blue 250:doi 41:IBM 39:of 438:: 392:. 356:. 328:: 326:}} 322:{{ 266:. 256:. 248:. 206:. 97:. 424:. 402:. 367:. 338:) 318:. 281:. 252:: 218:. 102:p 94:i 90:d 88:p 84:p 79:i 77:p 73:p 69:p 65:p

Index

scheduling discipline
network scheduler
Kang G. Shin
University of Michigan
Thomas J. Watson Research Center
IBM
random early detection
explicit congestion notification
network interface controller
queuing disciplines
traffic flows
Stochastic Fairness Queuing
bloom filter
hash table
penalty box
distributed denial-of-service
ALTQ
network scheduler
Linux
Linux kernel
"BLUE: A New Class of Active Queue Management Algorithms"
Proceedings IEEE INFOCOM 2001. Conference on Computer Communications. Twentieth Annual Joint Conference of the IEEE Computer and Communications Society (Cat. No.01CH37213)
CiteSeerX
10.1.1.11.4235
doi
10.1109/INFCOM.2001.916648
ISBN
978-0-7803-7016-6
S2CID
5902623

Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.