![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to transition.tex CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [CSRG BSD Unix] / 43BSDReno / contrib / isode-beta / doc / practical-osi1|
Return to transition.tex CVS log | Up to [CSRG BSD Unix] / 43BSDReno / contrib / isode-beta / doc / practical-osi1 |
1.1 root 1: % run this through LaTeX with the appropriate wrapper
2:
3: \dotopic{2}
4: \begin{bwslide}
5: \part {TRANSITION AND\\ COEXISTENCE\\ WITH TCP/IP}
6: \end{bwslide}
7: \doparts
8:
9:
10: \begin{bwslide}
11: \part* {OUTLINE}\bf
12:
13: \begin{description}
14: \item[PART I:] MOTIVATION
15:
16: \item[PART II:] BACKGROUND
17:
18: \item[PART III:] PROTOCOL-BASED APPROACHES
19:
20: \item[PART IV:] SERVICE-BASED APPROACHES
21:
22: \item[PART V:] EXAMPLES
23: \end{description}
24: \end{bwslide}
25:
26:
27: \begin{bwslide}
28: \part {MOTIVATION}\bf
29:
30: \begin{nrtc}
31: \item THERE ARE MANY TCP/IP NETWORKS TODAY; THERE WILL BE MORE TOMORROW
32:
33: \item BY THE TIME OSI BECOMES A WORTHWHILE OPERATIONAL ALTERNATIVE,
34: THERE WILL BE MANY MORE TCP/IP NETWORKS THAN THERE ARE TODAY!
35:
36: \item PROBLEM: HOW TO PROTECT INSTALLED BASE?
37:
38: \item PROBLEM: HOW TO TRANSITION GRACEFULLY?
39: \end{nrtc}
40: \end{bwslide}
41:
42:
43: \begin{bwslide}
44: \ctitle {GROWTH OF TCP/IP}
45:
46: \begin{nrtc}
47: \item SALES OF TCP/IP-BASED TECHNOLOGY
48: \begin{nrtc}
49: \item PARTICULARLY IN EUROPE
50: \end{nrtc}
51: CONTINUES TO GROW
52:
53: \item SEVERAL TECHNICAL AND MARKET ASPECTS CONTRIBUTE TO THIS PHENOMENA:
54: \begin{nrtc}
55: \item SUPERIORITY OF TCP/IP IN LOWER-LAYER CONNECTIVITY
56:
57: \item MATURITY OF TCP/IP PRODUCTS\\ (e.g., RANGE OF PLATFORMS)
58: \end{nrtc}
59:
60: \item ALTHOUGH OSI WILL DOMINATE, IT DOESN'T YET
61:
62: \item HENCE, TCP/IP IS BECOMING MORE FIRMLY ENTRENCHED
63: \end{nrtc}
64: \end{bwslide}
65:
66:
67: \begin{bwslide}
68: \ctitle {FEAR AND LOATHING IN THE MARKET}
69:
70: \begin{nrtc}
71: \item F.U.D. IN THE MARKETPLACE:
72: \begin{quote}\em
73: ``All marketing is fear, uncertainty, and doubt.''\\ \raggedleft
74: -- Einar Stefferud, Network Management Associates
75: \end{quote}
76:
77: \item WHAT THE VENDORS SAY:
78: \begin{quote}\em
79: ``$\ldots$ protect your investment while assuring a path to an OSI
80: future.''\\ \raggedleft
81: -- Vendor A
82: \end{quote}
83: AND
84: \begin{quote}\em
85: ``$\ldots$ plans for a smooth, painless guaranteed migration to OSI standards
86: as they are approved.''\\ \raggedleft
87: --Vendor B
88: \end{quote}
89: AND
90: \begin{quote}\em
91: ``Once you've scrapped your existing production networks,
92: come to us for OSI.
93: It will be wonderful!''\\ \raggedleft
94: --Vendor C
95: \end{quote}
96: \end{nrtc}
97: \end{bwslide}
98:
99:
100: \begin{bwslide}
101: \ctitle {THE SAD TRUTH}
102:
103: \begin{quote}\em
104: ``You can't win, and you can't quit, but you \underline{can} reduce the
105: pain.''\\ \raggedleft
106: -- Marshall Rose, NYSERNet, Inc.
107: \end{quote}
108: \end{bwslide}
109:
110:
111: \begin{bwslide}
112: \part {BACKGROUND}\bf
113:
114: \begin{nrtc}
115: \item CONCEPTS
116:
117: \item TERMINOLOGY
118:
119: \item HISTORY
120:
121: \item METRICS FOR COMPARISON
122: \end{nrtc}
123: \end{bwslide}
124:
125:
126: \begin{bwslide}
127: \ctitle {THE FUNDAMENTAL ASSUMPTION}
128:
129: \begin{nrtc}
130: \item TCP/IP IS HERE TODAY, WIDELY INSTALLED, AND USEFUL
131:
132: \item OSI WILL EVENTUALLY REPLACE TCP/IP AS THE OFF-THE-SHELF TECHNOLOGY FOR
133: BUILDING INTEROPERABLE SYSTMS
134:
135: \item BOTH WILL BE SIMULTANEOUSLY WIDESPREAD FOR QUITE SOME TIME
136: \begin{nrtc}
137: \item DURING WHICH OSI WILL GAIN DOMINANCE
138: \end{nrtc}
139: \end{nrtc}
140: \end{bwslide}
141:
142:
143: \begin{bwslide}
144: \part* {CONCEPTS}\bf
145:
146: \begin{nrtc}
147: \item TRANSITION:
148: \begin{nrtc}
149: \item TO MOVE FROM ONE PROTOCOL SUITE TO ANOTHER
150: \end{nrtc}
151:
152: \item COEXISTENCE:
153: \begin{nrtc}
154: \item TO LIVE TOGETHER WITHOUT HOSTILITY OR CONFLICT DESPITE
155: DIFFERENCES
156: \end{nrtc}
157:
158: \item MIGRATION:
159: \begin{nrtc}
160: \item TO MOVE BACK AND FORTH, AS THE SEASONS CHANGE
161: \end{nrtc}
162: \end{nrtc}
163: \end{bwslide}
164:
165:
166: \begin{bwslide}
167: \ctitle {MAPPINGS}
168:
169: \begin{nrtc}
170: \item TRANSITION AND COEXISTENCE CAN BE DESCRIBED BY THE MAPPINGS THEY
171: REQUIRE
172:
173: \item SOME MAPPINGS ARE SIMPLE
174: \begin{nrtc}
175: \item i.e., SYNTACTIC CHANGES
176: \end{nrtc}
177:
178: \item SOME MAPPINGS ARE COMPLEX
179: \begin{nrtc}
180: \item i.e., SEMANTIC CHANGES
181: \end{nrtc}
182:
183: \item THE MORE COMPLEX THE MAPPING, THE GREATER THE LOSS OF INFORMATION OR
184: INTENT
185: \end{nrtc}
186: \end{bwslide}
187:
188:
189: \begin{bwslide}
190: \part* {TERMINOLOGY}\bf
191:
192: \begin{nrtc}
193: \item WE'LL FAVOR OSI TERMINOLOGY, BUT STILL NEED SOME INTERNET (TCP/IP)
194: TERMINOLOGY
195:
196: \item TWO BASIC TERMS
197: \begin{nrtc}
198: \item GATEWAY: GENERIC TO ANY LEVEL, COMPLEX
199:
200: \item BRIDGE: GENERIC TO ANY LEVEL, SIMPLE
201: \end{nrtc}
202: \end{nrtc}
203: \end{bwslide}
204:
205:
206: \begin{bwslide}
207: \ctitle {SERVICE SEMANTICS}
208:
209: \begin{nrtc}
210: \item STORE-AND-FORWARD
211: \begin{nrtc}
212: \item SERVICE SEMANTICS CARRIED MULTI-HOP VIA FORWARDERS
213: \end{nrtc}
214:
215: \item END-TO-END
216: \begin{nrtc}
217: \item SERVICE SEMANTICS CARRIED FROM ORIGINATOR TO RECIPIENT
218:
219: \item MAY BE SUPPORTED BY AN UNDERYLING STORE-AND-FORWARD SERVICE
220: \end{nrtc}
221: \end{nrtc}
222: \end{bwslide}
223:
224:
225: \begin{bwslide}
226: \ctitle {SERVICE SEMANTICS (cont.)}
227:
228: \vskip.5in
229: \diagram[p]{figureT-3}
230: \end{bwslide}
231:
232:
233: \begin{bwslide}
234: \ctitle {PROTOCOL SUITE}
235:
236: \begin{nrtc}
237: \item A COLLECTION OF SERVICES AND PROTOCOLS RELATED:
238: \begin{nrtc}
239: \item ADMINISTRATIVELY, BY AN ORGANIZATION\\ (e.g., ISO/IEC); and,
240:
241: \item PHILOSOPHICALLY, BY A REFERENCE MODEL\\ (e.g., the OSIRM)
242: \end{nrtc}
243:
244: \item FOR OUR PURPOSES, THERE ARE ONLY TWO:
245: \begin{nrtc}
246: \item THE OSI SUITE OF PROTOCOLS
247:
248: \item THE INTERNET SUITE OF PROTOCOLS
249: \end{nrtc}
250: \end{nrtc}
251: \end{bwslide}
252:
253:
254: \begin{bwslide}
255: \ctitle {APPLICATIONS}
256:
257: \begin{nrtc}
258: \item APPLICATION CLASS
259: \begin{nrtc}
260: \item A SET OF APPLICATIONS RELATED TO A PARTICULAR ACTIVITY,
261: e.g., FILE TRANSFER, IRREGARDLESS OF PROTOCOL SUITE
262: \end{nrtc}
263:
264: \item APPLICATION INSTANCE
265: \begin{nrtc}
266: \item A MEMBER OF AN APPLICATION CLASS SPECIFIC TO A PARTICULAR
267: PROTOCOL SUITE, e.g., FTAM
268: \end{nrtc}
269: \end{nrtc}
270: \end{bwslide}
271:
272:
273: \begin{bwslide}
274: \part* {HISTORY}\bf
275:
276: \begin{nrtc}
277: \item A VERY BRIEF INTRODUCTION TO THE TWO PROTOCOL SUITES
278:
279: \item WE'LL ATTEMPT TO TAKE A NON-PARTISAN VIEW (ha!)
280: \end{nrtc}
281: \end{bwslide}
282:
283:
284: \begin{bwslide}
285: \ctitle {INTERNET SUITE}
286:
287: \begin{nrtc}
288: \item SPONSORED BY THE U.S.~DoD
289: \begin{nrtc}
290: \item GREW OUT OF EARLY (D)ARPA RESEARCH INTO SURVIVABLE NETWORKS
291: \end{nrtc}
292: BASIS FROM THE DoD INTERNET ARCHITECTURE MODEL
293:
294: \item SPECIFIED IN ``REQUEST FOR COMMENTS'' SERIES (RFCs) AND
295: U.S.~MILITARY STANDARDS (MILSTDs)
296:
297: \item CURRENT GENERATION PRIMARILY BASED ON
298: \begin{nrtc}
299: \item CONNECTION-ORIENTED TRANSPORT SERVICE,
300: PROVIDED BY THE TCP; AND,
301:
302: \item CONNECTIONLESS-MODE NETWORK SERVICE,
303: PROVIDED BY THE IP
304: \end{nrtc}
305:
306: \item MAJOR EMPHASIS ON CONNECTIVITY OF DIVERSE SUB-NETWORKS
307: \begin{nrtc}
308: \item EXCELLENT RESEARCH CONTINUES, TO THIS DAY, ON THESE ISSUES
309: \end{nrtc}
310: \end{nrtc}
311: \end{bwslide}
312:
313:
314: \begin{bwslide}
315: \ctitle {INTERNET SUITE (cont.)}
316:
317: \begin{nrtc}
318: \item SEVERAL PRODUCTION APPLICATIONS
319: \begin{nrtc}
320: \item SIMPLE MAIL TRANSFER PROTOCOL (SMTP)
321:
322: \item FILE TRANSFER PROTOCOL (FTP)
323:
324: \item TELNET (VIRTUAL TERMINAL PROTOCOL)
325:
326: \item DOMAIN NAME SYSTEM (DNS)
327: \end{nrtc}
328: ALL OF WHICH ARE RATHER SIMPLE
329:
330: \item APPLICATIONS CONTAIN THEIR OWN IMPLICIT SESSION AND PRESENTATION
331: MECHANISMS
332:
333: \item NOT SURPRISING, CONSIDERING THAT THESE APPLICATIONS ARE ALL BASED ON
334: 15~YEAR OLD MODELS!
335: \end{nrtc}
336: \end{bwslide}
337:
338:
339: \begin{bwslide}
340: \ctitle {INTERNET PROTOCOLS}
341:
342: \vskip.5in
343: \diagram[p]{figureT-4}
344: \end{bwslide}
345:
346:
347: \begin{bwslide}
348: \ctitle {OSI SUITE}
349:
350: \begin{nrtc}
351: \item SPONSORED BY THE INTERNATIONAL COMMUNITY
352: \begin{nrtc}
353: \item IN PARTICULAR THE ISO
354: \end{nrtc}
355: BASIS FROM THE OSI REFERENCE MODEL (OSIRM)
356:
357: \item SPECIFIED IN ``STANDARDS'' (ISO/IEC) AND RECOMMENDATIONS (CCITT)
358:
359: \item BASED ON
360: \begin{nrtc}
361: \item CONNECTION-ORIENTED TRANSPORT SERVICE,
362: PROVIDED BY ONE OF FIVE DIFFERENT TPs; DEPENDING ON
363:
364: \item THE NETWORK SERVICE AVAILABLE (CONS or CLNS)
365: \end{nrtc}
366:
367: \item DIFFICULT TO IDENTIFY THE ``MAJOR'' EMPHASIS
368: \end{nrtc}
369: \end{bwslide}
370:
371:
372: \begin{bwslide}
373: \ctitle {OSI SUITE (cont.)}
374:
375: \begin{nrtc}
376: \item SEVERAL INTERESTING APPLICATIONS
377: \begin{nrtc}
378: \item MESSAGE HANDLING SYSTEMS (MHS)
379:
380: \item FILE TRANSFER, ACCESS AND MANAGEMENT (FTAM)
381:
382: \item VIRTUAL TERMINAL (VT)
383:
384: \item DIRECTORY SERVICES (DS)
385: \end{nrtc}
386:
387: \item APPLICATIONS EVOLVING QUITE HEAVILY OVER THE LAST FEW YEARS
388:
389: \item MUCH MORE AMBITIOUS THAN THEIR INTERNET COUNTERPARTS
390: \end{nrtc}
391: \end{bwslide}
392:
393:
394: \begin{bwslide}
395: %%%\ctitle {OSI PROTOCOLS}
396:
397: %%%\vskip.25in
398: \diagram[p]{figureT-5}
399: \end{bwslide}
400:
401:
402: \begin{bwslide}
403: \ctitle {A BRIEF COMPARISON}
404:
405: \begin{nrtc}
406: \item NOTE THAT CONCERNS DIFFER
407: \begin{nrtc}
408: \item NETWORK USERS: APPLICATION-LEVEL FUNCTIONALITY
409:
410: \item NETWORK ADMINISTRATORS: NETWORK AND TRANSPORT ISSUES
411: \end{nrtc}
412:
413: \item FOR APPLICATIONS, ONCE IMPLEMENTED, THE OSI SUITE IS SUPERIOR
414:
415: \item FOR NETWORK/TRANSPORT ISSUES, AT PRESENT,
416: THE INTERNET SUITE IS SUPERIOR
417: \end{nrtc}
418: \end{bwslide}
419:
420:
421: \begin{bwslide}
422: \part* {METRICS FOR COMPARISON}\bf
423:
424: \begin{nrtc}
425: \item CAN JUDGE A TRANSITION/COEXISTENCE SCHEME USING DIFFERENT
426: CRITERIA
427:
428: \item THE FOUR WE'LL FOCUS ON ARE ALL SUBJECTIVE;
429: \begin{nrtc}
430: \item TECHNICAL SOLUTIONS DO NOT EXIST IN A VACUUM
431:
432: \item THEY MUST BE EVALUATED IN THE CONTEXT OF A TARGET ENVIRONMENT
433: \end{nrtc}
434: \end{nrtc}
435: \end{bwslide}
436:
437:
438: \begin{bwslide}
439: \ctitle {METRICS FOR COMPARISON (cont.)}
440:
441: \begin{nrtc}
442: \item PERFORMANCE:
443: \begin{nrtc}
444: \item THROUGHPUT, LATENCY
445:
446: \item EFFECT ON OTHER APPLICATIONS
447: \end{nrtc}
448:
449: \item FLEXIBILITY:
450: \begin{nrtc}
451: \item RANGE OF APPLICABILITY
452: \end{nrtc}
453: \end{nrtc}
454: \end{bwslide}
455:
456:
457: \begin{bwslide}
458: \ctitle {METRICS FOR COMPARISON (cont.)}
459:
460: \begin{nrtc}
461: \item TRANSPARENCY:
462: \begin{nrtc}
463: \item USAGE CONTINUITY
464:
465: \item SEAMLESS USER INTERFACE
466: \end{nrtc}
467:
468: \item AMENABILITY:
469: \begin{nrtc}
470: \item MANAGEABILITY
471: \end{nrtc}
472: \end{nrtc}
473: \end{bwslide}
474:
475:
476: \begin{bwslide}
477: \ctitle {SEVERAL CANDIDATES}
478:
479: \begin{nrtc}
480: \item PROTOCOL-BASED APPROACHES
481: \begin{nrtc}
482: \item DUAL STACK
483:
484: \item APPLICATION GATEWAYS
485:
486: \item TRANSPORT GATEWAYS
487: \end{nrtc}
488:
489: \item SERVICE-BASED APPROACHES
490: \begin{nrtc}
491: \item TRANSPORT-SERVICE BRIDGES
492:
493: \item NETWORK TUNNELS
494: \end{nrtc}
495:
496: \item NONE OF THESE TECHNIQUES ARE SPECIFIC TO THE PROBLEM OF
497: \begin{nrtc}
498: \item INTERNET $\mapsto$ OSI
499: \end{nrtc}
500: \end{nrtc}
501: \end{bwslide}
502:
503:
504: \begin{bwslide}
505: \part {PROTOCOL-BASED APPROACHES}\bf
506:
507: \begin{nrtc}
508: \item THE ``STANDARD'' METHODS USED TO INTERCONNECT DIFFERENT
509: PROTOCOL STACKS
510:
511: \item THESE EMPHASIZE THE PROTOCOLS IN EACH STACK
512:
513: \item HENCE THEY REINFORCE THE BOUNDARIES BETWEEN TCP/IP AND OSI
514: \end{nrtc}
515: \end{bwslide}
516:
517:
518: \begin{bwslide}
519: \part* {DUAL STACK}\bf
520:
521: \begin{nrtc}
522: \item PUT BOTH PROTOCOL SUITES IN ALL HOSTS
523:
524: \item WORKS WELL, IF YOU CAN CHANGE EVERYTHING ON THE NETWORK
525: \begin{quote}\em
526: ``Nice work, if you can get it.''\\ \raggedleft
527: -- Groucho Marx, Monkey Business, Paramount Pictures (1931)
528: \end{quote}
529: \end{nrtc}
530: \end{bwslide}
531:
532:
533: \begin{bwslide}
534: \ctitle {DUAL STACK (cont.)}
535:
536: \vskip.5in
537: \diagram[p]{figureT-1}
538: \end{bwslide}
539:
540:
541: \begin{bwslide}
542: \ctitle {TALKING TO UNI-STACK HOSTS}
543:
544: \begin{nrtc}
545: \item QUESTION: HOW TO DECIDE WHICH APPLICATION INSTANCE,
546: \begin{nrtc}
547: \item APPL-$\alpha$ OR APPL-$\gamma$,
548: \end{nrtc}
549: TO USE?
550:
551: \item TWO ANSWERS:
552: \begin{nrtc}
553: \item DEPEND ON THE USER TO KNOW AND INVOKE THE RIGHT PROGRAM
554:
555: \item DEVELOP A GENERIC APPLICATION WHICH SUPPORTS BOTH CLASSES
556: \end{nrtc}
557:
558: \item IN THE LATTER CASE, NEED AN UP-TO-DATE DIRECTORY TO DO THIS RELIABLY
559: \end{nrtc}
560: \end{bwslide}
561:
562:
563: \begin{bwslide}
564: \ctitle {GENERIC APPLICATION INSTANCE}
565:
566: \vskip.5in
567: \diagram[p]{figureT-6}
568: \end{bwslide}
569:
570:
571: \begin{bwslide}
572: \ctitle {AN IMPLEMENTATION OF DUAL-STACK}
573:
574: \begin{nrtc}
575: \item ENVIRONMENT: \unix/~SVR3 (STREAMS)
576:
577: \item ACCESS TO LOWER-LAYER PROTOCOLS VIA TRANSPORT LAYER INTERFACE (TLI)
578:
579: \item NOTE THAT ALTHOUGH TLI PROVIDES A UNIFORM INTERFACE,
580: IT DOES NOT PROVIDE A UNIFORM SERVICE:
581: \begin{nrtc}
582: \item PACKET- vs. STREAM-ORIENTATION
583:
584: \item GRACEFUL RELEASE
585:
586: \item EXPEDITED vs. URGENT DATA
587:
588: \item ADDRESSING
589: \end{nrtc}
590: \end{nrtc}
591: \end{bwslide}
592:
593:
594: \begin{bwslide}
595: \ctitle {GENERIC APPLICATION INSTANCE}
596:
597: \vskip.5in
598: \diagram[p]{figureT-11}
599: \end{bwslide}
600:
601:
602: \begin{bwslide}
603: \ctitle {SCORECARD}
604:
605: \begin{nrtc}
606: \item PERFORMANCE: NO DEGRADATION
607:
608: \item FLEXIBILITY: GOOD
609:
610: \item TRANSPARENCY:
611: \begin{nrtc}
612: \item ASSUMING REMOTE SYSTEM SUPPORTS AT LEAST ONE OF THE PROTOCOL
613: STACKS, THEN HIGH TRANSPARENCY BY USING COMMON SERVICE
614: INTERFACE
615: \end{nrtc}
616:
617: \item AMENABILITY:
618: \begin{nrtc}
619: \item BOTH END- AND INTERMEDIATE-SYSTEMS MUST RUN BOTH PROTOCOLS
620:
621: \item INTRODUCES ADMINISTRATIVE PROBLEMS AS THERE ARE NOW TWO
622: LOGICAL NETWORKS
623: \begin{nrtc}
624: \item MANAGEMENT OF BOTH \underline{PLUS} CONTENTION BETWEEN THEM
625: \end{nrtc}
626: \end{nrtc}
627: \end{nrtc}
628: \end{bwslide}
629:
630:
631: \begin{bwslide}
632: \part* {APPLICATION GATEWAYS}\bf
633:
634: \begin{nrtc}
635: \item A WELL-KNOWN, BUT LITTLE-UNDERSTOOD TECHNOLOGY
636: \begin{nrtc}
637: \item USED IN MESSAGE HANDLING QUITE A BIT
638: \end{nrtc}
639:
640: \item MOST ARE QUITE TERRIBLE
641: \begin{quote}\em
642: ``Sometimes when you try to turn an apple into an orange you get back a
643: lemon.''\\ \raggedleft
644: -- Michael Padlipsky, The Elements of Networking Style (1985)
645: \end{quote}
646: \end{nrtc}
647: \end{bwslide}
648:
649:
650: \begin{bwslide}
651: \ctitle {APPLICATION GATEWAYS (cont.)}
652:
653: \vskip.5in
654: \diagram[p]{figureT-2}
655: \end{bwslide}
656:
657:
658: \begin{bwslide}
659: \ctitle {IMPERFECT MAPPINGS}
660:
661: \begin{nrtc}
662: \item BECAUSE THEY ARE AT THE HIGHEST LAYER IN THE STACK,
663: APPLICATION GATEWAYS TEND TO PERFORM SEMANTIC MAPPINGS
664:
665: \item THESE ARE ACCOMPANIED BY A LOSS OF INFORMATION
666:
667: \item SOMETIMES THE LOSS IS ONLY ANNOYING
668: \begin{nrtc}
669: \item e.g., ``FUNNY LOOKING'' MAIL ADDRESSES
670: \end{nrtc}
671:
672: \item SOMETIMES THE LOSS IS CATASTROPHIC
673: \begin{nrtc}
674: \item e.g., ROUTING LOOPS
675: \end{nrtc}
676: \end{nrtc}
677: \end{bwslide}
678:
679:
680: \begin{bwslide}
681: \ctitle {AN IMPLEMENATION OF APPLICATION-GATEWAY}
682:
683: \begin{nrtc}
684: \item TWO KINDS OF IMPLEMENATIONS
685:
686: \item STAGING (TRUE STORE-AND-FORWARD):
687: \begin{nrtc}
688: \item TOP-LEVEL PROTOCOL TRANSACTIONS ARE GROUPED AT THE GATEWAY
689:
690: \item REQUIRES LOCAL STORAGE, BUT MAY PERMIT BETTER MAPPINGS
691: \end{nrtc}
692:
693: \item IN-SITU (VIRTUAL END-TO-END):
694: \begin{nrtc}
695: \item NO PROTOCOL TRANSACTIONS ARE GROUPED
696:
697: \item MAPPINGS ARE ``ON THE FLY''\\ (AND PERHAPS LESS PRECISE)
698:
699: \item END-TO-END RESPONSE IS FASTER
700: \end{nrtc}
701: \end{nrtc}
702: \end{bwslide}
703:
704:
705: \begin{bwslide}
706: \ctitle {INVOKING THE GATEWAY}
707:
708: \vskip1.5in
709: \begin{verbatim}
710: % ftp file-gateway
711: Name (file-gateway:asterix): obelix@osi-host
712: Password:
713: \end{verbatim}
714: \end{bwslide}
715:
716:
717: \begin{bwslide}
718: \ctitle {A STAGING IMPLEMENTATION}
719:
720: \vskip.5in
721: \diagram[p]{figureT-12}
722: \end{bwslide}
723:
724:
725: \begin{bwslide}
726: \ctitle {AN IN-SITU IMPLEMENTATION}
727:
728: \vskip.5in
729: \diagram[p]{figureT-13}
730: \end{bwslide}
731:
732:
733: \begin{bwslide}
734: \ctitle {SCORECARD}
735:
736: \begin{nrtc}
737: \item PERFORMANCE: USUALLY POOR, BUT ACCEPTABLE FOR STORE-AND-FORWARD
738: APPLICATIONS
739: \begin{nrtc}
740: \item TYPICALLY ALSO INTRODUCES ADDITIONAL NETWORK TRAFFIC
741: \end{nrtc}
742:
743: \item FLEXIBILITY: NONE; EACH A-GWY IS A SPECIAL-PURPOSE SOFTWARE BOX
744:
745: \item TRANSPARENCY:
746: \begin{nrtc}
747: \item TO SERVICE: OFTEN LOSES SIGNIFICANT FUNCTIONALITY
748:
749: \item TO USERS: POSSIBLE, BUT NOT LIKELY (e.g., IN AN FTAM/FTP A-GWY,
750: USERS EMBED HOSTNAMES IN FILENAMES)
751: \end{nrtc}
752:
753: \item AMENABILITY:
754: \begin{nrtc}
755: \item REQUIRES NO END-SYSTEM MODIFICATION
756:
757: \item MAY INTRODUCE ADMINISTRATIVE PROBLEMS
758: \end{nrtc}
759: \end{nrtc}
760: \end{bwslide}
761:
762:
763: \begin{bwslide}
764: \part* {TRANSPORT GATEWAYS}\bf
765:
766: \begin{nrtc}
767: \item IDEA: GATEWAY AT THE TRANSPORT LAYER SO AS TO AVOID NEEDING
768: MULTIPLE APPLICATION GATEWAYS
769: \begin{quote}\em
770: ``We could do it, but it would be wrong.''\\ \raggedleft
771: -- Richard Nixon, The Watergate Tapes (1974)
772: \end{quote}
773:
774: \item ALTHOUGH THE OSI (TP4) AND INTERNET (TCP) TRANSPORT PROTOCOLS DIFFER,
775: THE SERVICE IS QUITE SIMILAR
776:
777: \item HENCE, IT IS TECHNICALLY FEASIBLE TO PERFORM THE MAPPINGS
778: \begin{nrtc}
779: \item (ALTHOUGH IT'S A LOT OF HARD WORK)
780: \end{nrtc}
781: \end{nrtc}
782: \end{bwslide}
783:
784:
785: \begin{bwslide}
786: \ctitle {TRANSPORT GATEWAYS (cont.)}
787:
788: \vskip.5in
789: \diagram[p]{figureT-14}
790: \end{bwslide}
791:
792:
793: \begin{bwslide}
794: \ctitle {THE OBVIOUS QUESTION}
795:
796: \begin{nrtc}
797: \item WHAT APPLICATION DO YOU RUN WHEN USING THIS?
798: \begin{nrtc}
799: \item CAN'T RUN INTERNET APPLICATIONS IN THE OSI NETWORK,
800: SINCE THE TRANSPORT GATEWAY YIELDS OSI TRANSPORT SEMANTICS
801:
802: \item CAN'T RUN OSI APPLICATIONS IN THE INTERNET NETWORK,
803: SINCE THE TRANSPORT GATEWAY YIELDS INTERNET TRANSPORT SEMANTICS
804: \end{nrtc}
805:
806: \item THIS APPROACH FAILS BECAUSE IT PRESENTS DIFFERENT SERVICE SEMANTICS
807: IN EACH NETWORK
808: \end{nrtc}
809: \end{bwslide}
810:
811:
812: \begin{bwslide}
813: \part {SERVICE-BASED APPROACHES}\bf
814:
815: \begin{nrtc}
816: \item BY THE TIME OSI-BASED NETWORKS ARE TRULY WIDESPREAD,
817: TCP/IP-BASED NETWORKS WILL ALREADY OFFER A MIX OF SERVICES:
818: \begin{nrtc}
819: \item SUCH AS FTAM AND MHS, IN ADDITION TO FTP AND SMTP
820: \end{nrtc}
821:
822: \item IN OTHER WORDS, PERHAPS THE TRANSITION TO OSI BEGINS WITH NEW
823: APPLICATIONS ON HOSTS AND NO CHANGES TO THE NETWORK
824: \end{nrtc}
825: \end{bwslide}
826:
827:
828: \begin{bwslide}
829: \ctitle {WOULD THIS REALLY HAPPEN?}
830:
831: \begin{nrtc}
832: \item RECALL THAT USERS ARE INTERESTED IN \underline{SERVICES} NOT
833: \underline{PROTOCOLS}
834:
835: \item THE OSI APPLICATIONS ARE MUCH RICHER THAN THEIR INTERNET COUNTERPARTS
836:
837: \item IN CONTRAST, AT THE LOWER-LAYERS THE INTERNET SUITE ``WORKS BETTER''
838: \begin{nrtc}
839: \item AS SUCH, IT IS UNLIKELY TO BE REPLACED BY THE OSI LOWER-LAYERS
840: FOR QUITE SOME TIME
841: \end{nrtc}
842: \end{nrtc}
843: \end{bwslide}
844:
845:
846: \begin{bwslide}
847: \ctitle {OBSERVATION}
848:
849: \begin{nrtc}
850: \item GIVEN THE ABOVE ASSUMPTION, IT SHOULD BE NOTED THAT:
851: \begin{nrtc}
852: \item WE HAVE TWO COMMUNITIES USING THE SAME APPLICATIONS
853: (OSI), AND
854:
855: \item ONLY THE UNDERLYING ``TS-STACK'' WILL DIFFER BETWEEN THE TWO:
856: \begin{nrtc}
857: \item IN THE OSI COMMUNITY: TP4/CLNP/$\ldots$
858:
859: \item IN THE TCP COMMUNITY: RFC1006/TCP/IP/$\ldots$
860: \end{nrtc}
861: \end{nrtc}
862:
863: \item THIS LEADS US TO POSTULATE AN INTERESTING COEXISTENCE
864: STRATEGY:
865: \begin{nrtc}
866: \item LET'S RUN OSI APPLICATIONS, END-TO-END, BETWEEN THE TWO
867: \end{nrtc}
868:
869: \item IN A SENSE, THIS IS A HYBRID OF THE TWO PREVIOUS APPROACHES,
870: INTENDED TO MINIMIZE THE DISADVANTAGES OF EACH
871: \begin{nrtc}
872: \item SAME APPLICATION PROTOCOL,\\
873: BUT DIFFERENT UNDERYLING LAYERS
874: \end{nrtc}
875: \end{nrtc}
876: \end{bwslide}
877:
878:
879: \begin{bwslide}
880: \part* {TRANSPORT-SERVICE BRIDGES}\bf
881:
882: \begin{nrtc}
883: \item INTRODUCE A TRANSPORT ENTITY CALLED THE ``TS-BRIDGE''
884: \begin{quote}\em
885: ``Users are interested in services, not protocols.''\\ \raggedleft
886: -- Marshall Rose, NYSERNet, Inc.
887: \end{quote}
888:
889: \item THE TS-BRIDGE ``COPIES'' SERVICE PRIMITIVES FROM ONE TS-STACK TO THE
890: OTHER, e.g.:
891: \begin{nrtc}
892: \item UPON RECEIVING A T-CONNECT.INDICATION PRIMITIVE FROM ONE
893: TS-STACK,
894:
895: \item IT ISSUES A T-CONNECT.REQUEST PRIMITIVE TO THE OTHER TS-STACK
896: \end{nrtc}
897:
898: \item AS DISCUSSED EARLIER, THIS TECHNOLOGY IS USED FOR CONNECTIVITY BETWEEN
899: DIFFERENT OSI COMMUNITIES
900: \end{nrtc}
901: \end{bwslide}
902:
903:
904: \begin{bwslide}
905: \ctitle {TRANSPORT-SERVICE BRIDGES (cont.)}
906:
907: \vskip.5in
908: \diagram[p]{figureT-9}
909: \end{bwslide}
910:
911:
912: \begin{bwslide}
913: \ctitle {CONS vs. CLNS CONNECTIVITY}
914:
915: \vskip.5in
916: \diagram[p]{figureT-19}
917: \end{bwslide}
918:
919:
920: \begin{bwslide}
921: \ctitle {THE TS-BRIDGE AND THE OSI MODEL\\ (REVIEW)}
922:
923: \begin{nrtc}
924: \item THE TS-BRIDGE IS A LEVEL-FOUR ROUTER
925:
926: \item POTENTIAL PROBLEMS:
927: \begin{nrtc}
928: \item THE TS-BRIDGE MAINTAINS STATE AS TO THE EXISTING CONNECTIONS
929:
930: \item TWO CHECKSUMS, AND NEITHER REALLY END-TO-END
931:
932: \item \underline{MAY} THWART SOPHISTICATED BACK-PRESSURE TECHNIQUES
933: \end{nrtc}
934: \end{nrtc}
935: \end{bwslide}
936:
937:
938: \begin{bwslide}
939: \ctitle {AN IMPLEMENTATION OF THE TS-BRIDGE}
940:
941: \begin{nrtc}
942: \item FIRST DEMONSTRATION IN FEBRUARY, 1988
943: \begin{nrtc}
944: \item TP4/CLNP to RFC1006/TCP
945: \end{nrtc}
946:
947: \item ANOTHER IMPLEMENTATION IN EUROPE IS HANDLING
948: \begin{nrtc}
949: \item TP0/X.25 to RFC1006/TCP
950: \end{nrtc}
951: \end{nrtc}
952: \end{bwslide}
953:
954:
955: \begin{bwslide}
956: \ctitle {SCORECARD}
957:
958: \begin{nrtc}
959: \item PERFORMANCE: FAIR; WHEN TS-BRIDGE IS MADE INTO A KERNEL-RESIDENT
960: STREAMS MODULE IT SHOULD IMPROVE DRAMATICALLY
961:
962: \item FLEXIBILITY: HIGH; INDEPENDENT OF ANY APPLICATION
963:
964: \item TRANSPARENCY: TOTAL
965:
966: \item AMENABILITY:
967: \begin{nrtc}
968: \item TCP END-SYSTEMS MUST RUN ``NEW'' PROTOCOLS
969: \begin{nrtc}
970: \item BUT, NO MODIFICATIONS REQUIRED TO END-SYSTEM KERNELS
971: \end{nrtc}
972:
973: \item MAY INTRODUCE ADMINISTRATIVE PROBLEMS
974: \end{nrtc}
975: \end{nrtc}
976: \end{bwslide}
977:
978:
979: \begin{bwslide}
980: \part* {NETWORK TUNNELS}\bf
981:
982: \begin{nrtc}
983: \item IDEA: ENCAPSULATE CLNP INSIDE OF IP, TREATING IP AS SIMPLY A DATA LINK
984: PROTOCOL
985: \begin{quote}\em
986: ``Encapsulation complies with the layering concept, but violates the notion
987: of absolute levels.''\\ \raggedleft
988: -- Danny Cohen and Jon Postel, ``The ISO Reference Model and Other Protocol
989: Architectures'' (1983)
990: \end{quote}
991:
992: \item NS-TUNNEL PERFORMS AS A ROUTER, REMOVING ONE DATA LINK HEADER AND
993: ADDING ANOTHER
994:
995: \item METHOD SPECIFIED IN [RFC1070]
996:
997: \item ADDRESS MAPPINGS SPECIFIED IN [RFC1069]
998: \end{nrtc}
999: \end{bwslide}
1000:
1001:
1002: \begin{bwslide}
1003: \ctitle {TUNNELING}
1004:
1005: \vskip.5in
1006: \diagram[p]{figureT-18}
1007: \end{bwslide}
1008:
1009:
1010: \begin{bwslide}
1011: \ctitle {NETWORK TUNNELS}
1012:
1013: \vskip.5in
1014: \diagram[p]{figureT-10}
1015: \end{bwslide}
1016:
1017:
1018: \begin{bwslide}
1019: \ctitle {INTERESTING FEATURES}
1020:
1021: \begin{nrtc}
1022: \item NO STATE MAINTAINED BY NS-TUNNEL
1023:
1024: \item A TRUE END-TO-END CHECKSUM
1025: \end{nrtc}
1026: \end{bwslide}
1027:
1028:
1029: \begin{bwslide}
1030: \ctitle {POTENTIAL PROBLEMS}
1031:
1032: \begin{nrtc}
1033: \item REQUIRES COMMON HIGHER-LEVEL PROTOCOLS (TRANSPORT AND ABOVE) ON BOTH
1034: END-SYSTEMS, BUT DOES NOT REQUIRE ALL INTERVENING ROUTERS TO USE THE
1035: SAME NETWORK PROTOCOL
1036:
1037: \item THE TCP END-SYSTEM IMPLEMENTATION CHOICES ARE SIMILAR TO NETBIOS OVER
1038: TCP [RFC1001/1002]
1039: \end{nrtc}
1040: \end{bwslide}
1041:
1042:
1043: \begin{bwslide}
1044: \ctitle {AN IMPLEMENATION OF AN NS-TUNNEL}
1045:
1046: \begin{nrtc}
1047: \item HAVEN'T SEE ANY YET
1048: \begin{nrtc}
1049: \item BUT WILL BE IN 4.4BSD UNIX
1050: \end{nrtc}
1051:
1052: \item NEED A LOT OF CLNP-BASED NETWORKS BEFORE THIS IS OF USE
1053:
1054: \item SO THIS WILL HAPPEN AT THE END OF THE TRANSITION PERIOD
1055: \end{nrtc}
1056: \end{bwslide}
1057:
1058:
1059: \begin{bwslide}
1060: \ctitle {SCORECARD}
1061:
1062: \begin{nrtc}
1063: \item PERFORMANCE: NO WORSE THAN TYPICAL CLNP-ROUTER (AND PROBABLY A LOT
1064: BETTER TOO!)
1065:
1066: \item FLEXIBILITY: HIGH (INDEPENDENT OF ANY APPLICATION)
1067:
1068: \item TRANSPARENCY: TOTAL
1069:
1070: \item AMENABILITY: TCP END-SYSTEMS MUST RUN BOTH TRANSPORT PROTOCOLS
1071: \end{nrtc}
1072: \end{bwslide}
1073:
1074:
1075: \begin{bwslide}
1076: \part {EXAMPLES}\bf
1077:
1078: \begin{nrtc}
1079: \item DoD OSI IMPLEMENTATION PLAN
1080:
1081: \item GENERIC EXAMPLE
1082:
1083: \item CONCLUSIONS
1084: \end{nrtc}
1085: \end{bwslide}
1086:
1087:
1088: \begin{bwslide}
1089: \part* {DoD OSI\\ IMPLEMENTATION PLAN}\bf
1090:
1091: \begin{nrtc}
1092: \item IMPLEMENT CAPABILITY TO USE OSI IN DoD INTERNETWORK ENVIRONMENT
1093: \begin{nrtc}
1094: \item OSI-POSIX PROJECT
1095: \end{nrtc}
1096:
1097: \item PROVIDE THE CAPABILITY FOR DoD AND OSI PROTOCOLS TO INTEROPERATE
1098: \begin{nrtc}
1099: \item FTAM-FTP GATEWAY
1100:
1101: \item MHS-SMTP GATEWAY
1102: \end{nrtc}
1103: \end{nrtc}
1104: \end{bwslide}
1105:
1106:
1107: \begin{bwslide}
1108: \ctitle {OSI-POSIX PROJECT}
1109:
1110: \begin{nrtc}
1111: \item GOAL: ACCELLERATE THE UBIQUITY OF OSI
1112:
1113: \item APPROACH: OPENLY AVAILABLE, COMPLETE OSI IMPLEMENTATION FOR NEXT MAJOR
1114: RELEASE OF BERKELEY \unix/
1115:
1116: \item FOR MORE DETAILS:
1117: \begin{quote}
1118: OSI PROTOCOLS WITHIN AN OPENLY AVAILABLE, POSIX-CONFORMANT, BERKELEY UNIX
1119: ENVIRONMENT
1120: \end{quote}
1121: APPEARING IN ConneXions, OCTOBER, 1988
1122: \end{nrtc}
1123: \end{bwslide}
1124:
1125:
1126: \begin{bwslide}
1127: \diagram[p]{figureT-15}
1128: \end{bwslide}
1129:
1130:
1131: \begin{bwslide}
1132: \diagram[p]{figureT-16}
1133: \end{bwslide}
1134:
1135:
1136: \begin{bwslide}
1137: \part* {GENERIC EXAMPLE}\bf
1138:
1139: \begin{nrtc}
1140: \item TWO PRONGS:
1141: \begin{nrtc}
1142: \item FAVOR USE OF OSI APPLICATIONS OVER TCP ON LAN MESH
1143:
1144: \item LOCATE APPLICATION GATEWAYS AND A TS-BRIDGE ON ALL NODES
1145: WITH WAN ATTACHMENETS
1146: \end{nrtc}
1147:
1148: \item AWAIT OSI LOWER-LAYERS TO BECOME COMPETITIVE
1149: \end{nrtc}
1150: \end{bwslide}
1151:
1152:
1153: \begin{bwslide}
1154: \ctitle {GENERIC EXAMPLE (cont.)}
1155:
1156: \begin{nrtc}
1157: \item EACH ATTACHMENT LOCUS SHOULD SUPPORT COEXISTENCE SERVICES
1158:
1159: \item IF RESOURCES PERMIT, SELECT ONE OTHER SYSTEM TO SUPPORT THESE
1160: SERVICES FOR USE BY LOCAL UNI-STACK HOSTS
1161:
1162: \item THIS ``COVERS ALL BASES'' BY HANDLING ALL POSSIBLE OSI COMBINATIONS
1163: WITH A BIT OF EXTRA REDUNDANCY
1164:
1165: \item MIGHT REQUIRE A BIT OF SOPHISTICATED USE FROM THE DIRECTORY
1166: \end{nrtc}
1167: \end{bwslide}
1168:
1169:
1170: \begin{bwslide}
1171: \ctitle {A LAN OF MANY COLORS}
1172:
1173: \vskip.5in
1174: \diagram[p]{figureT-17}
1175: \end{bwslide}
1176:
1177:
1178: \begin{bwslide}
1179: \part* {CONCLUSIONS}\bf
1180:
1181: \begin{quote}\em
1182: ``Optimality differs according to context.''\\ \raggedleft
1183: -- Michael Padlipsky, The Elements of Networking Style (1985)
1184: \end{quote}
1185: \end{bwslide}
1186:
1187:
1188: \begin{bwslide}
1189: \ctitle {CONCLUSIONS (cont.)}
1190:
1191: \begin{nrtc}
1192: \item TCP/IP-BASED NETWORKS WILL OFFER OSI APPLICATIONS
1193:
1194: \item COEXISTENCE IN THE SHORT TERM:
1195: \begin{nrtc}
1196: \item TS-BRIDGE MINIMIZES SOFTWARE INVESTMENT
1197: \end{nrtc}
1198:
1199: \item COEXISTENCE IN THE LONG TERM:
1200: \begin{nrtc}
1201: \item NS-TUNNEL MAXIMIZES PERFORMANCE AND ROBUSTNESS
1202: \end{nrtc}
1203:
1204: \item IF/WHEN THERE ARE NO MORE TCP/IP-BASED NETWORKS, THEN THE
1205: COEXISTENCE PERIOD IS OVER, AND TRANSITION IS A NON-ISSUE!
1206: \end{nrtc}
1207: \end{bwslide}
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.