summaryrefslogtreecommitdiff
path: root/lib/tdb2/doc/design.lyx
blob: 0a1d6a14bca57bf0a9bcf02c1265301907cf1f90 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
#LyX 1.6.7 created this file. For more info see http://www.lyx.org/
\lyxformat 345
\begin_document
\begin_header
\textclass article
\use_default_options true
\language english
\inputencoding auto
\font_roman default
\font_sans default
\font_typewriter default
\font_default_family default
\font_sc false
\font_osf false
\font_sf_scale 100
\font_tt_scale 100

\graphics default
\paperfontsize default
\use_hyperref false
\papersize default
\use_geometry false
\use_amsmath 1
\use_esint 1
\cite_engine basic
\use_bibtopic false
\paperorientation portrait
\secnumdepth 3
\tocdepth 3
\paragraph_separation indent
\defskip medskip
\quotes_language english
\papercolumns 1
\papersides 1
\paperpagestyle default
\tracking_changes true
\output_changes true
\author ""
\author ""
\end_header

\begin_body

\begin_layout Title
TDB2: A Redesigning The Trivial DataBase
\end_layout

\begin_layout Author
Rusty Russell, IBM Corporation
\end_layout

\begin_layout Date
17-March-2011
\end_layout

\begin_layout Abstract
The Trivial DataBase on-disk format is 32 bits; with usage cases heading
 towards the 4G limit, that must change.
 This required breakage provides an opportunity to revisit TDB's other design
 decisions and reassess them.
\end_layout

\begin_layout Section
Introduction
\end_layout

\begin_layout Standard
The Trivial DataBase was originally written by Andrew Tridgell as a simple
 key/data pair storage system with the same API as dbm, but allowing multiple
 readers and writers while being small enough (< 1000 lines of C) to include
 in SAMBA.
 The simple design created in 1999 has proven surprisingly robust and performant
, used in Samba versions 3 and 4 as well as numerous other projects.
 Its useful life was greatly increased by the (backwards-compatible!) addition
 of transaction support in 2005.
\end_layout

\begin_layout Standard
The wider variety and greater demands of TDB-using code has lead to some
 organic growth of the API, as well as some compromises on the implementation.
 None of these, by themselves, are seen as show-stoppers, but the cumulative
 effect is to a loss of elegance over the initial, simple TDB implementation.
 Here is a table of the approximate number of lines of implementation code
 and number of API functions at the end of each year:
\end_layout

\begin_layout Standard
\begin_inset Tabular
<lyxtabular version="3" rows="12" columns="3">
<features>
<column alignment="center" valignment="top" width="0">
<column alignment="center" valignment="top" width="0">
<column alignment="center" valignment="top" width="0">
<row>
<cell alignment="center" valignment="top" topline="true" bottomline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
Year End
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" bottomline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
API Functions
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" bottomline="true" leftline="true" rightline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
Lines of C Code Implementation
\end_layout

\end_inset
</cell>
</row>
<row>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
1999
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
13
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
1195
\end_layout

\end_inset
</cell>
</row>
<row>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2000
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
24
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
1725
\end_layout

\end_inset
</cell>
</row>
<row>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2001
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
32
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2228
\end_layout

\end_inset
</cell>
</row>
<row>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2002
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
35
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2481
\end_layout

\end_inset
</cell>
</row>
<row>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2003
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
35
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2552
\end_layout

\end_inset
</cell>
</row>
<row>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2004
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
40
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2584
\end_layout

\end_inset
</cell>
</row>
<row>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2005
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
38
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2647
\end_layout

\end_inset
</cell>
</row>
<row>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2006
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
52
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
3754
\end_layout

\end_inset
</cell>
</row>
<row>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2007
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
66
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
4398
\end_layout

\end_inset
</cell>
</row>
<row>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2008
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
71
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" leftline="true" rightline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
4768
\end_layout

\end_inset
</cell>
</row>
<row>
<cell alignment="center" valignment="top" topline="true" bottomline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
2009
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" bottomline="true" leftline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
73
\end_layout

\end_inset
</cell>
<cell alignment="center" valignment="top" topline="true" bottomline="true" leftline="true" rightline="true" usebox="none">
\begin_inset Text

\begin_layout Plain Layout
5715
\end_layout

\end_inset
</cell>
</row>
</lyxtabular>

\end_inset


\end_layout

\begin_layout Standard
This review is an attempt to catalog and address all the known issues with
 TDB and create solutions which address the problems without significantly
 increasing complexity; all involved are far too aware of the dangers of
 second system syndrome in rewriting a successful project like this.
\end_layout

\begin_layout Section
API Issues
\end_layout

\begin_layout Subsection
tdb_open_ex Is Not Expandable
\end_layout

\begin_layout Standard
The tdb_open() call was expanded to tdb_open_ex(), which added an optional
 hashing function and an optional logging function argument.
 Additional arguments to open would require the introduction of a tdb_open_ex2
 call etc.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\begin_inset CommandInset label
LatexCommand label
name "attributes"

\end_inset


\end_layout

\begin_layout Standard
tdb_open() will take a linked-list of attributes:
\end_layout

\begin_layout LyX-Code
enum tdb_attribute {
\end_layout

\begin_layout LyX-Code
    TDB_ATTRIBUTE_LOG = 0,
\end_layout

\begin_layout LyX-Code
    TDB_ATTRIBUTE_HASH = 1
\end_layout

\begin_layout LyX-Code
};
\end_layout

\begin_layout LyX-Code
struct tdb_attribute_base {
\end_layout

\begin_layout LyX-Code
    enum tdb_attribute attr;
\end_layout

\begin_layout LyX-Code
    union tdb_attribute *next;
\end_layout

\begin_layout LyX-Code
};
\end_layout

\begin_layout LyX-Code
struct tdb_attribute_log {
\end_layout

\begin_layout LyX-Code
    struct tdb_attribute_base base; /* .attr = TDB_ATTRIBUTE_LOG */
\end_layout

\begin_layout LyX-Code
    tdb_log_func log_fn;
\end_layout

\begin_layout LyX-Code
    void *log_private;
\end_layout

\begin_layout LyX-Code
};
\end_layout

\begin_layout LyX-Code
struct tdb_attribute_hash {
\end_layout

\begin_layout LyX-Code
    struct tdb_attribute_base base; /* .attr = TDB_ATTRIBUTE_HASH */
\end_layout

\begin_layout LyX-Code
    tdb_hash_func hash_fn;
\end_layout

\begin_layout LyX-Code
    void *hash_private;
\end_layout

\begin_layout LyX-Code
};
\end_layout

\begin_layout LyX-Code
union tdb_attribute {
\end_layout

\begin_layout LyX-Code
    struct tdb_attribute_base base;
\end_layout

\begin_layout LyX-Code
    struct tdb_attribute_log log;
\end_layout

\begin_layout LyX-Code
    struct tdb_attribute_hash hash;
\end_layout

\begin_layout LyX-Code
};
\end_layout

\begin_layout Standard
This allows future attributes to be added, even if this expands the size
 of the union.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
tdb_traverse Makes Impossible Guarantees
\end_layout

\begin_layout Standard
tdb_traverse (and tdb_firstkey/tdb_nextkey) predate transactions, and it
 was thought that it was important to guarantee that all records which exist
 at the start and end of the traversal would be included, and no record
 would be included twice.
\end_layout

\begin_layout Standard
This adds complexity (see
\begin_inset CommandInset ref
LatexCommand ref
reference "Reliable-Traversal-Adds"

\end_inset

) and does not work anyway for records which are altered (in particular,
 those which are expanded may be effectively deleted and re-added behind
 the traversal).
\end_layout

\begin_layout Subsubsection
\begin_inset CommandInset label
LatexCommand label
name "traverse-Proposed-Solution"

\end_inset

Proposed Solution
\end_layout

\begin_layout Standard
Abandon the guarantee.
 You will see every record if no changes occur during your traversal, otherwise
 you will see some subset.
 You can prevent changes by using a transaction or the locking API.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
 Delete-during-traverse will still delete every record, too (assuming no
 other changes).
\end_layout

\begin_layout Subsection
Nesting of Transactions Is Fraught
\end_layout

\begin_layout Standard
TDB has alternated between allowing nested transactions and not allowing
 them.
 Various paths in the Samba codebase assume that transactions will nest,
 and in a sense they can: the operation is only committed to disk when the
 outer transaction is committed.
 There are two problems, however:
\end_layout

\begin_layout Enumerate
Canceling the inner transaction will cause the outer transaction commit
 to fail, and will not undo any operations since the inner transaction began.
 This problem is soluble with some additional internal code.
\end_layout

\begin_layout Enumerate
An inner transaction commit can be cancelled by the outer transaction.
 This is desirable in the way which Samba's database initialization code
 uses transactions, but could be a surprise to any users expecting a successful
 transaction commit to expose changes to others.
\end_layout

\begin_layout Standard
The current solution is to specify the behavior at tdb_open(), with the
 default currently that nested transactions are allowed.
 This flag can also be changed at runtime.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
Given the usage patterns, it seems that the
\begin_inset Quotes eld
\end_inset

least-surprise
\begin_inset Quotes erd
\end_inset

 behavior of disallowing nested transactions should become the default.
 Additionally, it seems the outer transaction is the only code which knows
 whether inner transactions should be allowed, so a flag to indicate this
 could be added to tdb_transaction_start.
 However, this behavior can be simulated with a wrapper which uses tdb_add_flags
() and tdb_remove_flags(), so the API should not be expanded for this relatively
-obscure case.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete; the nesting flag has been removed.
\end_layout

\begin_layout Subsection
Incorrect Hash Function is Not Detected
\end_layout

\begin_layout Standard
tdb_open_ex() allows the calling code to specify a different hash function
 to use, but does not check that all other processes accessing this tdb
 are using the same hash function.
 The result is that records are missing from tdb_fetch().
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
The header should contain an example hash result (eg.
 the hash of 0xdeadbeef), and tdb_open_ex() should check that the given
 hash function produces the same answer, or fail the tdb_open call.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
tdb_set_max_dead/TDB_VOLATILE Expose Implementation
\end_layout

\begin_layout Standard
In response to scalability issues with the free list (
\begin_inset CommandInset ref
LatexCommand ref
reference "TDB-Freelist-Is"

\end_inset

) two API workarounds have been incorporated in TDB: tdb_set_max_dead()
 and the TDB_VOLATILE flag to tdb_open.
 The latter actually calls the former with an argument of
\begin_inset Quotes eld
\end_inset

5
\begin_inset Quotes erd
\end_inset

.
\end_layout

\begin_layout Standard
This code allows deleted records to accumulate without putting them in the
 free list.
 On delete we iterate through each chain and free them in a batch if there
 are more than max_dead entries.
 These are never otherwise recycled except as a side-effect of a tdb_repack.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
With the scalability problems of the freelist solved, this API can be removed.
 The TDB_VOLATILE flag may still be useful as a hint that store and delete
 of records will be at least as common as fetch in order to allow some internal
 tuning, but initially will become a no-op.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
 Unknown flags cause tdb_open() to fail as well, so they can be detected
 at runtime.
\end_layout

\begin_layout Subsection
\begin_inset CommandInset label
LatexCommand label
name "TDB-Files-Cannot"

\end_inset

TDB Files Cannot Be Opened Multiple Times In The Same Process
\end_layout

\begin_layout Standard
No process can open the same TDB twice; we check and disallow it.
 This is an unfortunate side-effect of fcntl locks, which operate on a per-file
 rather than per-file-descriptor basis, and do not nest.
 Thus, closing any file descriptor on a file clears all the locks obtained
 by this process, even if they were placed using a different file descriptor!
\end_layout

\begin_layout Standard
Note that even if this were solved, deadlock could occur if operations were
 nested: this is a more manageable programming error in most cases.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
We could lobby POSIX to fix the perverse rules, or at least lobby Linux
 to violate them so that the most common implementation does not have this
 restriction.
 This would be a generally good idea for other fcntl lock users.
\end_layout

\begin_layout Standard
Samba uses a wrapper which hands out the same tdb_context to multiple callers
 if this happens, and does simple reference counting.
 We should do this inside the tdb library, which already emulates lock nesting
 internally; it would need to recognize when deadlock occurs within a single
 process.
 This would create a new failure mode for tdb operations (while we currently
 handle locking failures, they are impossible in normal use and a process
 encountering them can do little but give up).
\end_layout

\begin_layout Standard
I do not see benefit in an additional tdb_open flag to indicate whether
 re-opening is allowed, as though there may be some benefit to adding a
 call to detect when a tdb_context is shared, to allow other to create such
 an API.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
TDB API Is Not POSIX Thread-safe
\end_layout

\begin_layout Standard
The TDB API uses an error code which can be queried after an operation to
 determine what went wrong.
 This programming model does not work with threads, unless specific additional
 guarantees are given by the implementation.
 In addition, even otherwise-independent threads cannot open the same TDB
 (as in
\begin_inset CommandInset ref
LatexCommand ref
reference "TDB-Files-Cannot"

\end_inset

).
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
Reachitecting the API to include a tdb_errcode pointer would be a great
 deal of churn, but fortunately most functions return 0 on success and -1
 on error: we can change these to return 0 on success and a negative error
 code on error, and the API remains similar to previous.
 The tdb_fetch, tdb_firstkey and tdb_nextkey functions need to take a TDB_DATA
 pointer and return an error code.
 It is also simpler to have tdb_nextkey replace its key argument in place,
 freeing up any old .dptr.
\end_layout

\begin_layout Standard
Internal locking is required to make sure that fcntl locks do not overlap
 between threads, and also that the global list of tdbs is maintained.
\end_layout

\begin_layout Standard
The aim is that building tdb with -DTDB_PTHREAD will result in a pthread-safe
 version of the library, and otherwise no overhead will exist.
 Alternatively, a hooking mechanism similar to that proposed for
\begin_inset CommandInset ref
LatexCommand ref
reference "Proposed-Solution-locking-hook"

\end_inset

 could be used to enable pthread locking at runtime.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Incomplete; API has been changed but thread safety has not been implemented.
\end_layout

\begin_layout Subsection
*_nonblock Functions And *_mark Functions Expose Implementation
\end_layout

\begin_layout Standard
CTDB
\begin_inset Foot
status collapsed

\begin_layout Plain Layout
Clustered TDB, see http://ctdb.samba.org
\end_layout

\end_inset

 wishes to operate on TDB in a non-blocking manner.
 This is currently done as follows:
\end_layout

\begin_layout Enumerate
Call the _nonblock variant of an API function (eg.
 tdb_lockall_nonblock).
 If this fails:
\end_layout

\begin_layout Enumerate
Fork a child process, and wait for it to call the normal variant (eg.
 tdb_lockall).
\end_layout

\begin_layout Enumerate
If the child succeeds, call the _mark variant to indicate we already have
 the locks (eg.
 tdb_lockall_mark).
\end_layout

\begin_layout Enumerate
Upon completion, tell the child to release the locks (eg.
 tdb_unlockall).
\end_layout

\begin_layout Enumerate
Indicate to tdb that it should consider the locks removed (eg.
 tdb_unlockall_mark).
\end_layout

\begin_layout Standard
There are several issues with this approach.
 Firstly, adding two new variants of each function clutters the API for
 an obscure use, and so not all functions have three variants.
 Secondly, it assumes that all paths of the functions ask for the same locks,
 otherwise the parent process will have to get a lock which the child doesn't
 have under some circumstances.
 I don't believe this is currently the case, but it constrains the implementatio
n.

\end_layout

\begin_layout Subsubsection
\begin_inset CommandInset label
LatexCommand label
name "Proposed-Solution-locking-hook"

\end_inset

Proposed Solution
\end_layout

\begin_layout Standard
Implement a hook for locking methods, so that the caller can control the
 calls to create and remove fcntl locks.
 In this scenario, ctdbd would operate as follows:
\end_layout

\begin_layout Enumerate
Call the normal API function, eg tdb_lockall().
\end_layout

\begin_layout Enumerate
When the lock callback comes in, check if the child has the lock.
 Initially, this is always false.
 If so, return 0.
 Otherwise, try to obtain it in non-blocking mode.
 If that fails, return EWOULDBLOCK.
\end_layout

\begin_layout Enumerate
Release locks in the unlock callback as normal.
\end_layout

\begin_layout Enumerate
If tdb_lockall() fails, see if we recorded a lock failure; if so, call the
 child to repeat the operation.
\end_layout

\begin_layout Enumerate
The child records what locks it obtains, and returns that information to
 the parent.
\end_layout

\begin_layout Enumerate
When the child has succeeded, goto 1.
\end_layout

\begin_layout Standard
This is flexible enough to handle any potential locking scenario, even when
 lock requirements change.
 It can be optimized so that the parent does not release locks, just tells
 the child which locks it doesn't need to obtain.
\end_layout

\begin_layout Standard
It also keeps the complexity out of the API, and in ctdbd where it is needed.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Incomplete.
\end_layout

\begin_layout Subsection
tdb_chainlock Functions Expose Implementation
\end_layout

\begin_layout Standard
tdb_chainlock locks some number of records, including the record indicated
 by the given key.
 This gave atomicity guarantees; no-one can start a transaction, alter,
 read or delete that key while the lock is held.
\end_layout

\begin_layout Standard
It also makes the same guarantee for any other key in the chain, which is
 an internal implementation detail and potentially a cause for deadlock.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
None.
 It would be nice to have an explicit single entry lock which effected no
 other keys.
 Unfortunately, this won't work for an entry which doesn't exist.
 Thus while chainlock may be implemented more efficiently for the existing
 case, it will still have overlap issues with the non-existing case.
 So it is best to keep the current (lack of) guarantee about which records
 will be effected to avoid constraining our implementation.
\end_layout

\begin_layout Subsection
Signal Handling is Not Race-Free
\end_layout

\begin_layout Standard
The tdb_setalarm_sigptr() call allows the caller's signal handler to indicate
 that the tdb locking code should return with a failure, rather than trying
 again when a signal is received (and errno == EAGAIN).
 This is usually used to implement timeouts.
\end_layout

\begin_layout Standard
Unfortunately, this does not work in the case where the signal is received
 before the tdb code enters the fcntl() call to place the lock: the code
 will sleep within the fcntl() code, unaware that the signal wants it to
 exit.
 In the case of long timeouts, this does not happen in practice.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
The locking hooks proposed in
\begin_inset CommandInset ref
LatexCommand ref
reference "Proposed-Solution-locking-hook"

\end_inset

 would allow the user to decide on whether to fail the lock acquisition
 on a signal.
 This allows the caller to choose their own compromise: they could narrow
 the race by checking immediately before the fcntl call.
\begin_inset Foot
status collapsed

\begin_layout Plain Layout
It may be possible to make this race-free in some implementations by having
 the signal handler alter the struct flock to make it invalid.
 This will cause the fcntl() lock call to fail with EINVAL if the signal
 occurs before the kernel is entered, otherwise EAGAIN.
\end_layout

\end_inset


\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Incomplete.
\end_layout

\begin_layout Subsection
The API Uses Gratuitous Typedefs, Capitals
\end_layout

\begin_layout Standard
typedefs are useful for providing source compatibility when types can differ
 across implementations, or arguably in the case of function pointer definitions
 which are hard for humans to parse.
 Otherwise it is simply obfuscation and pollutes the namespace.
\end_layout

\begin_layout Standard
Capitalization is usually reserved for compile-time constants and macros.
\end_layout

\begin_layout Description
TDB_CONTEXT There is no reason to use this over 'struct tdb_context'; the
 definition isn't visible to the API user anyway.
\end_layout

\begin_layout Description
TDB_DATA There is no reason to use this over struct TDB_DATA; the struct
 needs to be understood by the API user.
\end_layout

\begin_layout Description
struct
\begin_inset space ~
\end_inset

TDB_DATA This would normally be called 'struct tdb_data'.
\end_layout

\begin_layout Description
enum
\begin_inset space ~
\end_inset

TDB_ERROR Similarly, this would normally be enum tdb_error.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
None.
 Introducing lower case variants would please pedants like myself, but if
 it were done the existing ones should be kept.
 There is little point forcing a purely cosmetic change upon tdb users.
\end_layout

\begin_layout Subsection
\begin_inset CommandInset label
LatexCommand label
name "tdb_log_func-Doesnt-Take"

\end_inset

tdb_log_func Doesn't Take The Private Pointer
\end_layout

\begin_layout Standard
For API compatibility reasons, the logging function needs to call tdb_get_loggin
g_private() to retrieve the pointer registered by the tdb_open_ex for logging.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
It should simply take an extra argument, since we are prepared to break
 the API/ABI.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
Various Callback Functions Are Not Typesafe
\end_layout

\begin_layout Standard
The callback functions in tdb_set_logging_function (after
\begin_inset CommandInset ref
LatexCommand ref
reference "tdb_log_func-Doesnt-Take"

\end_inset

 is resolved), tdb_parse_record, tdb_traverse, tdb_traverse_read and tdb_check
 all take void * and must internally convert it to the argument type they
 were expecting.
\end_layout

\begin_layout Standard
If this type changes, the compiler will not produce warnings on the callers,
 since it only sees void *.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
With careful use of macros, we can create callback functions which give
 a warning when used on gcc and the types of the callback and its private
 argument differ.
 Unsupported compilers will not give a warning, which is no worse than now.
 In addition, the callbacks become clearer, as they need not use void *
 for their parameter.
\end_layout

\begin_layout Standard
See CCAN's typesafe_cb module at http://ccan.ozlabs.org/info/typesafe_cb.html
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
TDB_CLEAR_IF_FIRST Must Be Specified On All Opens, tdb_reopen_all Problematic
\end_layout

\begin_layout Standard
The TDB_CLEAR_IF_FIRST flag to tdb_open indicates that the TDB file should
 be cleared if the caller discovers it is the only process with the TDB
 open.
 However, if any caller does not specify TDB_CLEAR_IF_FIRST it will not
 be detected, so will have the TDB erased underneath them (usually resulting
 in a crash).
\end_layout

\begin_layout Standard
There is a similar issue on fork(); if the parent exits (or otherwise closes
 the tdb) before the child calls tdb_reopen_all() to establish the lock
 used to indicate the TDB is opened by someone, a TDB_CLEAR_IF_FIRST opener
 at that moment will believe it alone has opened the TDB and will erase
 it.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
Remove TDB_CLEAR_IF_FIRST.
 Other workarounds are possible, but see
\begin_inset CommandInset ref
LatexCommand ref
reference "TDB_CLEAR_IF_FIRST-Imposes-Performance"

\end_inset

.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
Extending The Header Is Difficult
\end_layout

\begin_layout Standard
We have reserved (zeroed) words in the TDB header, which can be used for
 future features.
 If the future features are compulsory, the version number must be updated
 to prevent old code from accessing the database.
 But if the future feature is optional, we have no way of telling if older
 code is accessing the database or not.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
The header should contain a
\begin_inset Quotes eld
\end_inset

format variant
\begin_inset Quotes erd
\end_inset

 value (64-bit).
 This is divided into two 32-bit parts:
\end_layout

\begin_layout Enumerate
The lower part reflects the format variant understood by code accessing
 the database.
\end_layout

\begin_layout Enumerate
The upper part reflects the format variant you must understand to write
 to the database (otherwise you can only open for reading).
\end_layout

\begin_layout Standard
The latter field can only be written at creation time, the former should
 be written under the OPEN_LOCK when opening the database for writing, if
 the variant of the code is lower than the current lowest variant.
\end_layout

\begin_layout Standard
This should allow backwards-compatible features to be added, and detection
 if older code (which doesn't understand the feature) writes to the database.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
Record Headers Are Not Expandible
\end_layout

\begin_layout Standard
If we later want to add (say) checksums on keys and data, it would require
 another format change, which we'd like to avoid.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
We often have extra padding at the tail of a record.
 If we ensure that the first byte (if any) of this padding is zero, we will
 have a way for future changes to detect code which doesn't understand a
 new format: the new code would write (say) a 1 at the tail, and thus if
 there is no tail or the first byte is 0, we would know the extension is
 not present on that record.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
TDB Does Not Use Talloc
\end_layout

\begin_layout Standard
Many users of TDB (particularly Samba) use the talloc allocator, and thus
 have to wrap TDB in a talloc context to use it conveniently.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
The allocation within TDB is not complicated enough to justify the use of
 talloc, and I am reluctant to force another (excellent) library on TDB
 users.
 Nonetheless a compromise is possible.
 An attribute (see
\begin_inset CommandInset ref
LatexCommand ref
reference "attributes"

\end_inset

) can be added later to tdb_open() to provide an alternate allocation mechanism,
 specifically for talloc but usable by any other allocator (which would
 ignore the
\begin_inset Quotes eld
\end_inset

context
\begin_inset Quotes erd
\end_inset

 argument).
\end_layout

\begin_layout Standard
This would form a talloc heirarchy as expected, but the caller would still
 have to attach a destructor to the tdb context returned from tdb_open to
 close it.
 All TDB_DATA fields would be children of the tdb_context, and the caller
 would still have to manage them (using talloc_free() or talloc_steal()).
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Deferred.
\end_layout

\begin_layout Section
Performance And Scalability Issues
\end_layout

\begin_layout Subsection
\begin_inset CommandInset label
LatexCommand label
name "TDB_CLEAR_IF_FIRST-Imposes-Performance"

\end_inset

TDB_CLEAR_IF_FIRST Imposes Performance Penalty
\end_layout

\begin_layout Standard
When TDB_CLEAR_IF_FIRST is specified, a 1-byte read lock is placed at offset
 4 (aka.
 the ACTIVE_LOCK).
 While these locks never conflict in normal tdb usage, they do add substantial
 overhead for most fcntl lock implementations when the kernel scans to detect
 if a lock conflict exists.
 This is often a single linked list, making the time to acquire and release
 a fcntl lock O(N) where N is the number of processes with the TDB open,
 not the number actually doing work.
\end_layout

\begin_layout Standard
In a Samba server it is common to have huge numbers of clients sitting idle,
 and thus they have weaned themselves off the TDB_CLEAR_IF_FIRST flag.
\begin_inset Foot
status collapsed

\begin_layout Plain Layout
There is a flag to tdb_reopen_all() which is used for this optimization:
 if the parent process will outlive the child, the child does not need the
 ACTIVE_LOCK.
 This is a workaround for this very performance issue.
\end_layout

\end_inset


\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
Remove the flag.
 It was a neat idea, but even trivial servers tend to know when they are
 initializing for the first time and can simply unlink the old tdb at that
 point.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
TDB Files Have a 4G Limit
\end_layout

\begin_layout Standard
This seems to be becoming an issue (so much for
\begin_inset Quotes eld
\end_inset

trivial
\begin_inset Quotes erd
\end_inset

!), particularly for ldb.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
A new, incompatible TDB format which uses 64 bit offsets internally rather
 than 32 bit as now.
 For simplicity of endian conversion (which TDB does on the fly if required),
 all values will be 64 bit on disk.
 In practice, some upper bits may be used for other purposes, but at least
 56 bits will be available for file offsets.
\end_layout

\begin_layout Standard
tdb_open() will automatically detect the old version, and even create them
 if TDB_VERSION6 is specified to tdb_open.
\end_layout

\begin_layout Standard
32 bit processes will still be able to access TDBs larger than 4G (assuming
 that their off_t allows them to seek to 64 bits), they will gracefully
 fall back as they fail to mmap.
 This can happen already with large TDBs.
\end_layout

\begin_layout Standard
Old versions of tdb will fail to open the new TDB files (since 28 August
 2009, commit 398d0c29290: prior to that any unrecognized file format would
 be erased and initialized as a fresh tdb!)
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
TDB Records Have a 4G Limit
\end_layout

\begin_layout Standard
This has not been a reported problem, and the API uses size_t which can
 be 64 bit on 64 bit platforms.
 However, other limits may have made such an issue moot.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
Record sizes will be 64 bit, with an error returned on 32 bit platforms
 which try to access such records (the current implementation would return
 TDB_ERR_OOM in a similar case).
 It seems unlikely that 32 bit keys will be a limitation, so the implementation
 may not support this (see
\begin_inset CommandInset ref
LatexCommand ref
reference "sub:Records-Incur-A"

\end_inset

).
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
Hash Size Is Determined At TDB Creation Time
\end_layout

\begin_layout Standard
TDB contains a number of hash chains in the header; the number is specified
 at creation time, and defaults to 131.
 This is such a bottleneck on large databases (as each hash chain gets quite
 long), that LDB uses 10,000 for this hash.
 In general it is impossible to know what the 'right' answer is at database
 creation time.
\end_layout

\begin_layout Subsubsection
\begin_inset CommandInset label
LatexCommand label
name "sub:Hash-Size-Solution"

\end_inset

Proposed Solution
\end_layout

\begin_layout Standard
After comprehensive performance testing on various scalable hash variants
\begin_inset Foot
status collapsed

\begin_layout Plain Layout
http://rusty.ozlabs.org/?p=89 and http://rusty.ozlabs.org/?p=94 This was annoying
 because I was previously convinced that an expanding tree of hashes would
 be very close to optimal.
\end_layout

\end_inset

, it became clear that it is hard to beat a straight linear hash table which
 doubles in size when it reaches saturation.
 Unfortunately, altering the hash table introduces serious locking complications
: the entire hash table needs to be locked to enlarge the hash table, and
 others might be holding locks.
 Particularly insidious are insertions done under tdb_chainlock.
\end_layout

\begin_layout Standard
Thus an expanding layered hash will be used: an array of hash groups, with
 each hash group exploding into pointers to lower hash groups once it fills,
 turning into a hash tree.
 This has implications for locking: we must lock the entire group in case
 we need to expand it, yet we don't know how deep the tree is at that point.
\end_layout

\begin_layout Standard
Note that bits from the hash table entries should be stolen to hold more
 hash bits to reduce the penalty of collisions.
 We can use the otherwise-unused lower 3 bits.
 If we limit the size of the database to 64 exabytes, we can use the top
 8 bits of the hash entry as well.
 These 11 bits would reduce false positives down to 1 in 2000 which is more
 than we need: we can use one of the bits to indicate that the extra hash
 bits are valid.
 This means we can choose not to re-hash all entries when we expand a hash
 group; simply use the next bits we need and mark them invalid.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
\begin_inset CommandInset label
LatexCommand label
name "TDB-Freelist-Is"

\end_inset

TDB Freelist Is Highly Contended
\end_layout

\begin_layout Standard
TDB uses a single linked list for the free list.
 Allocation occurs as follows, using heuristics which have evolved over
 time:
\end_layout

\begin_layout Enumerate
Get the free list lock for this whole operation.
\end_layout

\begin_layout Enumerate
Multiply length by 1.25, so we always over-allocate by 25%.
\end_layout

\begin_layout Enumerate
Set the slack multiplier to 1.
\end_layout

\begin_layout Enumerate
Examine the current freelist entry: if it is > length but < the current
 best case, remember it as the best case.
\end_layout

\begin_layout Enumerate
Multiply the slack multiplier by 1.05.
\end_layout

\begin_layout Enumerate
If our best fit so far is less than length * slack multiplier, return it.
 The slack will be turned into a new free record if it's large enough.
\end_layout

\begin_layout Enumerate
Otherwise, go onto the next freelist entry.
\end_layout

\begin_layout Standard
Deleting a record occurs as follows:
\end_layout

\begin_layout Enumerate
Lock the hash chain for this whole operation.
\end_layout

\begin_layout Enumerate
Walk the chain to find the record, keeping the prev pointer offset.
\end_layout

\begin_layout Enumerate
If max_dead is non-zero:
\end_layout

\begin_deeper
\begin_layout Enumerate
Walk the hash chain again and count the dead records.
\end_layout

\begin_layout Enumerate
If it's more than max_dead, bulk free all the dead ones (similar to steps
 4 and below, but the lock is only obtained once).
\end_layout

\begin_layout Enumerate
Simply mark this record as dead and return.

\end_layout

\end_deeper
\begin_layout Enumerate
Get the free list lock for the remainder of this operation.
\end_layout

\begin_layout Enumerate
\begin_inset CommandInset label
LatexCommand label
name "right-merging"

\end_inset

Examine the following block to see if it is free; if so, enlarge the current
 block and remove that block from the free list.
 This was disabled, as removal from the free list was O(entries-in-free-list).
\end_layout

\begin_layout Enumerate
Examine the preceeding block to see if it is free: for this reason, each
 block has a 32-bit tailer which indicates its length.
 If it is free, expand it to cover our new block and return.
\end_layout

\begin_layout Enumerate
Otherwise, prepend ourselves to the free list.
\end_layout

\begin_layout Standard
Disabling right-merging (step
\begin_inset CommandInset ref
LatexCommand ref
reference "right-merging"

\end_inset

) causes fragmentation; the other heuristics proved insufficient to address
 this, so the final answer to this was that when we expand the TDB file
 inside a transaction commit, we repack the entire tdb.
\end_layout

\begin_layout Standard
The single list lock limits our allocation rate; due to the other issues
 this is not currently seen as a bottleneck.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
The first step is to remove all the current heuristics, as they obviously
 interact, then examine them once the lock contention is addressed.
\end_layout

\begin_layout Standard
The free list must be split to reduce contention.
 Assuming perfect free merging, we can at most have 1 free list entry for
 each entry.
 This implies that the number of free lists is related to the size of the
 hash table, but as it is rare to walk a large number of free list entries
 we can use far fewer, say 1/32 of the number of hash buckets.
\end_layout

\begin_layout Standard
It seems tempting to try to reuse the hash implementation which we use for
 records here, but we have two ways of searching for free entries: for allocatio
n we search by size (and possibly zone) which produces too many clashes
 for our hash table to handle well, and for coalescing we search by address.
 Thus an array of doubly-linked free lists seems preferable.
\end_layout

\begin_layout Standard
There are various benefits in using per-size free lists (see
\begin_inset CommandInset ref
LatexCommand ref
reference "sub:TDB-Becomes-Fragmented"

\end_inset

) but it's not clear this would reduce contention in the common case where
 all processes are allocating/freeing the same size.
 Thus we almost certainly need to divide in other ways: the most obvious
 is to divide the file into zones, and using a free list (or table of free
 lists) for each.
 This approximates address ordering.
\end_layout

\begin_layout Standard
Unfortunately it is difficult to know what heuristics should be used to
 determine zone sizes, and our transaction code relies on being able to
 create a
\begin_inset Quotes eld
\end_inset

recovery area
\begin_inset Quotes erd
\end_inset

 by simply appending to the file (difficult if it would need to create a
 new zone header).
 Thus we use a linked-list of free tables; currently we only ever create
 one, but if there is more than one we choose one at random to use.
 In future we may use heuristics to add new free tables on contention.
 We only expand the file when all free tables are exhausted.
\end_layout

\begin_layout Standard
The basic algorithm is as follows.
 Freeing is simple:
\end_layout

\begin_layout Enumerate
Identify the correct free list.
\end_layout

\begin_layout Enumerate
Lock the corresponding list.
\end_layout

\begin_layout Enumerate
Re-check the list (we didn't have a lock, sizes could have changed): relock
 if necessary.
\end_layout

\begin_layout Enumerate
Place the freed entry in the list.
\end_layout

\begin_layout Standard
Allocation is a little more complicated, as we perform delayed coalescing
 at this point:
\end_layout

\begin_layout Enumerate
Pick a free table; usually the previous one.
\end_layout

\begin_layout Enumerate
Lock the corresponding list.
\end_layout

\begin_layout Enumerate
If the top entry is -large enough, remove it from the list and return it.
\end_layout

\begin_layout Enumerate
Otherwise, coalesce entries in the list.If there was no entry large enough,
 unlock the list and try the next largest list
\end_layout

\begin_layout Enumerate
If no list has an entry which meets our needs, try the next free table.
\end_layout

\begin_layout Enumerate
If no zone satisfies, expand the file.
\end_layout

\begin_layout Standard
This optimizes rapid insert/delete of free list entries by not coalescing
 them all the time..
 First-fit address ordering ordering seems to be fairly good for keeping
 fragmentation low (see
\begin_inset CommandInset ref
LatexCommand ref
reference "sub:TDB-Becomes-Fragmented"

\end_inset

).
 Note that address ordering does not need a tailer to coalesce, though if
 we needed one we could have one cheaply: see
\begin_inset CommandInset ref
LatexCommand ref
reference "sub:Records-Incur-A"

\end_inset

.

\end_layout

\begin_layout Standard
Each free entry has the free table number in the header: less than 255.
 It also contains a doubly-linked list for easy deletion.
\end_layout

\begin_layout Subsection
\begin_inset CommandInset label
LatexCommand label
name "sub:TDB-Becomes-Fragmented"

\end_inset

TDB Becomes Fragmented
\end_layout

\begin_layout Standard
Much of this is a result of allocation strategy
\begin_inset Foot
status collapsed

\begin_layout Plain Layout
The Memory Fragmentation Problem: Solved? Johnstone & Wilson 1995 ftp://ftp.cs.ute
xas.edu/pub/garbage/malloc/ismm98.ps
\end_layout

\end_inset

 and deliberate hobbling of coalescing; internal fragmentation (aka overallocati
on) is deliberately set at 25%, and external fragmentation is only cured
 by the decision to repack the entire db when a transaction commit needs
 to enlarge the file.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
The 25% overhead on allocation works in practice for ldb because indexes
 tend to expand by one record at a time.
 This internal fragmentation can be resolved by having an
\begin_inset Quotes eld
\end_inset

expanded
\begin_inset Quotes erd
\end_inset

 bit in the header to note entries that have previously expanded, and allocating
 more space for them.
\end_layout

\begin_layout Standard
There are is a spectrum of possible solutions for external fragmentation:
 one is to use a fragmentation-avoiding allocation strategy such as best-fit
 address-order allocator.
 The other end of the spectrum would be to use a bump allocator (very fast
 and simple) and simply repack the file when we reach the end.
\end_layout

\begin_layout Standard
There are three problems with efficient fragmentation-avoiding allocators:
 they are non-trivial, they tend to use a single free list for each size,
 and there's no evidence that tdb allocation patterns will match those recorded
 for general allocators (though it seems likely).
\end_layout

\begin_layout Standard
Thus we don't spend too much effort on external fragmentation; we will be
 no worse than the current code if we need to repack on occasion.
 More effort is spent on reducing freelist contention, and reducing overhead.
\end_layout

\begin_layout Subsection
\begin_inset CommandInset label
LatexCommand label
name "sub:Records-Incur-A"

\end_inset

Records Incur A 28-Byte Overhead
\end_layout

\begin_layout Standard
Each TDB record has a header as follows:
\end_layout

\begin_layout LyX-Code
struct tdb_record {
\end_layout

\begin_layout LyX-Code
        tdb_off_t next; /* offset of the next record in the list */
\end_layout

\begin_layout LyX-Code
        tdb_len_t rec_len; /* total byte length of record */
\end_layout

\begin_layout LyX-Code
        tdb_len_t key_len; /* byte length of key */
\end_layout

\begin_layout LyX-Code
        tdb_len_t data_len; /* byte length of data */
\end_layout

\begin_layout LyX-Code
        uint32_t full_hash; /* the full 32 bit hash of the key */
\end_layout

\begin_layout LyX-Code
        uint32_t magic;   /* try to catch errors */
\end_layout

\begin_layout LyX-Code
        /* the following union is implied:
\end_layout

\begin_layout LyX-Code
                union {
\end_layout

\begin_layout LyX-Code
                        char record[rec_len];
\end_layout

\begin_layout LyX-Code
                        struct {
\end_layout

\begin_layout LyX-Code
                                char key[key_len];
\end_layout

\begin_layout LyX-Code
                                char data[data_len];
\end_layout

\begin_layout LyX-Code
                        }
\end_layout

\begin_layout LyX-Code
                        uint32_t totalsize; (tailer)
\end_layout

\begin_layout LyX-Code
                }
\end_layout

\begin_layout LyX-Code
        */
\end_layout

\begin_layout LyX-Code
};
\end_layout

\begin_layout Standard
Naively, this would double to a 56-byte overhead on a 64 bit implementation.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
We can use various techniques to reduce this for an allocated block:
\end_layout

\begin_layout Enumerate
The 'next' pointer is not required, as we are using a flat hash table.
\end_layout

\begin_layout Enumerate
'rec_len' can instead be expressed as an addition to key_len and data_len
 (it accounts for wasted or overallocated length in the record).
 Since the record length is always a multiple of 8, we can conveniently
 fit it in 32 bits (representing up to 35 bits).
\end_layout

\begin_layout Enumerate
'key_len' and 'data_len' can be reduced.
 I'm unwilling to restrict 'data_len' to 32 bits, but instead we can combine
 the two into one 64-bit field and using a 5 bit value which indicates at
 what bit to divide the two.
 Keys are unlikely to scale as fast as data, so I'm assuming a maximum key
 size of 32 bits.
\end_layout

\begin_layout Enumerate
'full_hash' is used to avoid a memcmp on the
\begin_inset Quotes eld
\end_inset

miss
\begin_inset Quotes erd
\end_inset

 case, but this is diminishing returns after a handful of bits (at 10 bits,
 it reduces 99.9% of false memcmp).
 As an aside, as the lower bits are already incorporated in the hash table
 resolution, the upper bits should be used here.
 Note that it's not clear that these bits will be a win, given the extra
 bits in the hash table itself (see
\begin_inset CommandInset ref
LatexCommand ref
reference "sub:Hash-Size-Solution"

\end_inset

).
\end_layout

\begin_layout Enumerate
'magic' does not need to be enlarged: it currently reflects one of 5 values
 (used, free, dead, recovery, and unused_recovery).
 It is useful for quick sanity checking however, and should not be eliminated.
\end_layout

\begin_layout Enumerate
'tailer' is only used to coalesce free blocks (so a block to the right can
 find the header to check if this block is free).
 This can be replaced by a single 'free' bit in the header of the following
 block (and the tailer only exists in free blocks).
\begin_inset Foot
status collapsed

\begin_layout Plain Layout
This technique from Thomas Standish.
 Data Structure Techniques.
 Addison-Wesley, Reading, Massachusetts, 1980.
\end_layout

\end_inset

 The current proposed coalescing algorithm doesn't need this, however.
\end_layout

\begin_layout Standard
This produces a 16 byte used header like this:
\end_layout

\begin_layout LyX-Code
struct tdb_used_record {
\end_layout

\begin_layout LyX-Code
        uint32_t used_magic : 16,
\end_layout

\begin_layout LyX-Code

\end_layout

\begin_layout LyX-Code
                 key_data_divide: 5,
\end_layout

\begin_layout LyX-Code
                 top_hash: 11;
\end_layout

\begin_layout LyX-Code
        uint32_t extra_octets;
\end_layout

\begin_layout LyX-Code
        uint64_t key_and_data_len;
\end_layout

\begin_layout LyX-Code
};
\end_layout

\begin_layout Standard
And a free record like this:
\end_layout

\begin_layout LyX-Code
struct tdb_free_record {
\end_layout

\begin_layout LyX-Code
        uint64_t free_magic: 8,
\end_layout

\begin_layout LyX-Code
                   prev : 56;
\end_layout

\begin_layout LyX-Code

\end_layout

\begin_layout LyX-Code
        uint64_t free_table: 8,
\end_layout

\begin_layout LyX-Code
                 total_length : 56
\end_layout

\begin_layout LyX-Code
        uint64_t next;;
\end_layout

\begin_layout LyX-Code
};
\end_layout

\begin_layout Standard
Note that by limiting valid offsets to 56 bits, we can pack everything we
 need into 3 64-byte words, meaning our minimum record size is 8 bytes.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
Transaction Commit Requires 4 fdatasync
\end_layout

\begin_layout Standard
The current transaction algorithm is:
\end_layout

\begin_layout Enumerate
write_recovery_data();
\end_layout

\begin_layout Enumerate
sync();
\end_layout

\begin_layout Enumerate
write_recovery_header();
\end_layout

\begin_layout Enumerate
sync();
\end_layout

\begin_layout Enumerate
overwrite_with_new_data();
\end_layout

\begin_layout Enumerate
sync();
\end_layout

\begin_layout Enumerate
remove_recovery_header();
\end_layout

\begin_layout Enumerate
sync();
\end_layout

\begin_layout Standard
On current ext3, each sync flushes all data to disk, so the next 3 syncs
 are relatively expensive.
 But this could become a performance bottleneck on other filesystems such
 as ext4.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
Neil Brown points out that this is overzealous, and only one sync is needed:
\end_layout

\begin_layout Enumerate
Bundle the recovery data, a transaction counter and a strong checksum of
 the new data.
\end_layout

\begin_layout Enumerate
Strong checksum that whole bundle.
\end_layout

\begin_layout Enumerate
Store the bundle in the database.
\end_layout

\begin_layout Enumerate
Overwrite the oldest of the two recovery pointers in the header (identified
 using the transaction counter) with the offset of this bundle.
\end_layout

\begin_layout Enumerate
sync.
\end_layout

\begin_layout Enumerate
Write the new data to the file.
\end_layout

\begin_layout Standard
Checking for recovery means identifying the latest bundle with a valid checksum
 and using the new data checksum to ensure that it has been applied.
 This is more expensive than the current check, but need only be done at
 open.
 For running databases, a separate header field can be used to indicate
 a transaction in progress; we need only check for recovery if this is set.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Deferred.
\end_layout

\begin_layout Subsection
\begin_inset CommandInset label
LatexCommand label
name "sub:TDB-Does-Not"

\end_inset

TDB Does Not Have Snapshot Support
\end_layout

\begin_layout Subsubsection
Proposed SolutionNone.
 At some point you say
\begin_inset Quotes eld
\end_inset

use a real database
\begin_inset Quotes erd
\end_inset

 (but see
\begin_inset CommandInset ref
LatexCommand ref
reference "replay-attribute"

\end_inset

).
\end_layout

\begin_layout Standard
But as a thought experiment, if we implemented transactions to only overwrite
 free entries (this is tricky: there must not be a header in each entry
 which indicates whether it is free, but use of presence in metadata elsewhere),
 and a pointer to the hash table, we could create an entirely new commit
 without destroying existing data.
 Then it would be easy to implement snapshots in a similar way.
\end_layout

\begin_layout Standard
This would not allow arbitrary changes to the database, such as tdb_repack
 does, and would require more space (since we have to preserve the current
 and future entries at once).
 If we used hash trees rather than one big hash table, we might only have
 to rewrite some sections of the hash, too.
\end_layout

\begin_layout Standard
We could then implement snapshots using a similar method, using multiple
 different hash tables/free tables.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Deferred.
\end_layout

\begin_layout Subsection
Transactions Cannot Operate in Parallel
\end_layout

\begin_layout Standard
This would be useless for ldb, as it hits the index records with just about
 every update.
 It would add significant complexity in resolving clashes, and cause the
 all transaction callers to write their code to loop in the case where the
 transactions spuriously failed.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
None (but see
\begin_inset CommandInset ref
LatexCommand ref
reference "replay-attribute"

\end_inset

).
 We could solve a small part of the problem by providing read-only transactions.
 These would allow one write transaction to begin, but it could not commit
 until all r/o transactions are done.
 This would require a new RO_TRANSACTION_LOCK, which would be upgraded on
 commit.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Deferred.
\end_layout

\begin_layout Subsection
Default Hash Function Is Suboptimal
\end_layout

\begin_layout Standard
The Knuth-inspired multiplicative hash used by tdb is fairly slow (especially
 if we expand it to 64 bits), and works best when the hash bucket size is
 a prime number (which also means a slow modulus).
 In addition, it is highly predictable which could potentially lead to a
 Denial of Service attack in some TDB uses.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
The Jenkins lookup3 hash
\begin_inset Foot
status open

\begin_layout Plain Layout
http://burtleburtle.net/bob/c/lookup3.c
\end_layout

\end_inset

 is a fast and superbly-mixing hash.
 It's used by the Linux kernel and almost everything else.
 This has the particular properties that it takes an initial seed, and produces
 two 32 bit hash numbers, which we can combine into a 64-bit hash.
\end_layout

\begin_layout Standard
The seed should be created at tdb-creation time from some random source,
 and placed in the header.
 This is far from foolproof, but adds a little bit of protection against
 hash bombing.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
\begin_inset CommandInset label
LatexCommand label
name "Reliable-Traversal-Adds"

\end_inset

Reliable Traversal Adds Complexity
\end_layout

\begin_layout Standard
We lock a record during traversal iteration, and try to grab that lock in
 the delete code.
 If that grab on delete fails, we simply mark it deleted and continue onwards;
 traversal checks for this condition and does the delete when it moves off
 the record.
\end_layout

\begin_layout Standard
If traversal terminates, the dead record may be left indefinitely.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
Remove reliability guarantees; see
\begin_inset CommandInset ref
LatexCommand ref
reference "traverse-Proposed-Solution"

\end_inset

.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Complete.
\end_layout

\begin_layout Subsection
Fcntl Locking Adds Overhead
\end_layout

\begin_layout Standard
Placing a fcntl lock means a system call, as does removing one.
 This is actually one reason why transactions can be faster (everything
 is locked once at transaction start).
 In the uncontended case, this overhead can theoretically be eliminated.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
None.
\end_layout

\begin_layout Standard
We tried this before with spinlock support, in the early days of TDB, and
 it didn't make much difference except in manufactured benchmarks.
\end_layout

\begin_layout Standard
We could use spinlocks (with futex kernel support under Linux), but it means
 that we lose automatic cleanup when a process dies with a lock.
 There is a method of auto-cleanup under Linux, but it's not supported by
 other operating systems.
 We could reintroduce a clear-if-first-style lock and sweep for dead futexes
 on open, but that wouldn't help the normal case of one concurrent opener
 dying.
 Increasingly elaborate repair schemes could be considered, but they require
 an ABI change (everyone must use them) anyway, so there's no need to do
 this at the same time as everything else.
\end_layout

\begin_layout Subsection
Some Transactions Don't Require Durability
\end_layout

\begin_layout Standard
Volker points out that gencache uses a CLEAR_IF_FIRST tdb for normal (fast)
 usage, and occasionally empties the results into a transactional TDB.
 This kind of usage prioritizes performance over durability: as long as
 we are consistent, data can be lost.
\end_layout

\begin_layout Standard
This would be more neatly implemented inside tdb: a
\begin_inset Quotes eld
\end_inset

soft
\begin_inset Quotes erd
\end_inset

 transaction commit (ie.
 syncless) which meant that data may be reverted on a crash.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\end_layout

\begin_layout Standard
None.
\end_layout

\begin_layout Standard
Unfortunately any transaction scheme which overwrites old data requires
 a sync before that overwrite to avoid the possibility of corruption.
\end_layout

\begin_layout Standard
It seems possible to use a scheme similar to that described in
\begin_inset CommandInset ref
LatexCommand ref
reference "sub:TDB-Does-Not"

\end_inset

,where transactions are committed without overwriting existing data, and
 an array of top-level pointers were available in the header.
 If the transaction is
\begin_inset Quotes eld
\end_inset

soft
\begin_inset Quotes erd
\end_inset

 then we would not need a sync at all: existing processes would pick up
 the new hash table and free list and work with that.
\end_layout

\begin_layout Standard
At some later point, a sync would allow recovery of the old data into the
 free lists (perhaps when the array of top-level pointers filled).
 On crash, tdb_open() would examine the array of top levels, and apply the
 transactions until it encountered an invalid checksum.
\end_layout

\begin_layout Subsection
Tracing Is Fragile, Replay Is External
\end_layout

\begin_layout Standard
The current TDB has compile-time-enabled tracing code, but it often breaks
 as it is not enabled by default.
 In a similar way, the ctdb code has an external wrapper which does replay
 tracing so it can coordinate cluster-wide transactions.
\end_layout

\begin_layout Subsubsection
Proposed Solution
\begin_inset CommandInset label
LatexCommand label
name "replay-attribute"

\end_inset


\end_layout

\begin_layout Standard
Tridge points out that an attribute can be later added to tdb_open (see

\begin_inset CommandInset ref
LatexCommand ref
reference "attributes"

\end_inset

) to provide replay/trace hooks, which could become the basis for this and
 future parallel transactions and snapshot support.
\end_layout

\begin_layout Subsubsection
Status
\end_layout

\begin_layout Standard
Deferred.
\end_layout

\end_body
\end_document