1202:
administrator will claim 100% uptime. However, given the true definition of availability, the system will be approximately 99.9% available, or three nines (8751 hours of available time out of 8760 hours per non-leap year). Also, systems experiencing performance problems are often deemed partially or entirely unavailable by users, even when the systems are continuing to function. Similarly, unavailability of select application functions might go unnoticed by administrators yet be devastating to users – a true availability measure is holistic.
183:
1377:
the total number of components in the system. x is the number of components used to stress the system. N-1 means the model is stressed by evaluating performance with all possible combinations where one component is faulted. N-2 means the model is stressed by evaluating performance with all possible combinations where two component are faulted simultaneously.
1329:
Active redundancy is used in complex systems to achieve high availability with no performance decline. Multiple items of the same kind are incorporated into a design that includes a method to detect failure and automatically reconfigure the system to bypass failed items using a voting scheme. This is
1184:
documents. The use of the "nines" has been called into question, since it does not appropriately reflect that the impact of unavailability varies with its time of occurrence. For large amounts of 9s, the "unavailability" index (measure of downtime rather than uptime) is easier to handle. For example,
1015:
Similarly, percentages ending in a 5 have conventional names, traditionally the number of nines, then "five", so 99.95% is "three nines five", abbreviated 3N5. This is casually referred to as "three and a half nines", but this is incorrect: a 5 is only a factor of 2, while a 9 is a factor of 10, so a
751:
or customer. The subject of the terms is thus important here: whether the focus of a discussion is the server hardware, server OS, functional service, software service/process, or similar, it is only if there is a single, consistent subject of the discussion that the words uptime and availability can
259:
or system configuration changes that only take effect upon a reboot. In general, scheduled downtime is usually the result of some logical, management-initiated event. Unscheduled downtime events typically arise from some physical event, such as a hardware or software failure or environmental anomaly.
80:
The importance of network resilience is continuously increasing, as communication networks are becoming a fundamental component in the operation of critical infrastructures. Consequently, recent efforts focus on interpreting and improving network and computing resilience with applications to critical
1376:
is used to evaluate the theoretical reliability for large systems. The outcome of this kind of model is used to evaluate different design options. A model of the entire system is created, and the model is stressed by removing components. Redundancy simulation involves the N-x criteria. N represents
1201:
Availability measurement is subject to some degree of interpretation. A system that has been up for 365 days in a non-leap year might have been eclipsed by a network failure that lasted for 9 hours during a peak usage period; the user community will see the system as unavailable, whereas the system
1321:
Passive redundancy is used to achieve high availability by including enough excess capacity in the design to accommodate a performance decline. The simplest example is a boat with two separate engines driving two separate propellers. The boat continues toward its destination despite failure of a
1299:
inherently have more potential failure points and are more difficult to implement correctly. While some analysts would put forth the theory that the most highly available systems adhere to a simple architecture (a single, high quality, multi-purpose physical system with comprehensive internal
76:
and challenges to normal operation." Threats and challenges for services can range from simple misconfiguration over large scale natural disasters to targeted attacks. As such, network resilience touches a very wide range of topics. In order to increase the resilience of a given communication
1205:
Availability must be measured to be determined, ideally with comprehensive monitoring tools ("instrumentation") that are themselves highly available. If there is a lack of instrumentation, systems supporting high volume transaction processing throughout the day and night, such as credit card
1055:): 99.95% availability is 3.3 nines, not 3.5 nines. More simply, going from 99.9% availability to 99.95% availability is a factor of 2 (0.1% to 0.05% unavailability), but going from 99.95% to 99.99% availability is a factor of 5 (0.05% to 0.01% unavailability), over twice as much.
81:
infrastructures. As an example, one can consider as a resilience objective the provisioning of services over the network, instead of the services of the network itself. This may require coordinated response from both the network and from the services running on top of the network.
1300:
hardware redundancy), this architecture suffers from the requirement that the entire system must be brought down for patching and operating system upgrades. More advanced system designs allow for systems to be patched and upgraded without compromising service availability (see
1317:
is used to create systems with high levels of availability (e.g. aircraft flight computers). In this case it is required to have high levels of failure detectability and avoidance of common cause failures. Two kinds of redundancy are passive redundancy and active redundancy.
46:
refers to the ability of the user community to obtain a service or good, access the system, whether to submit new work, update or alter existing work, or collect the results of previous work. If a user cannot access the system, it is – from the user's point of view –
312:
often refer to monthly downtime or availability in order to calculate service credits to match monthly billing cycles. The following table shows the translation from a given availability percentage to the corresponding amount of time a system would be unavailable.
286:
Many computing sites exclude scheduled downtime from availability calculations, assuming that it has little or no impact upon the computing user community. By doing this, they can claim to have phenomenally high availability, which might give the illusion of
764:
allows approximately 5 minutes of downtime per year. Variants can be derived by multiplying or dividing by 10: 4 nines is 50 minutes and 3 nines is 500 minutes. In the opposite direction, 6 nines is 0.5 minutes (30 sec) and 7 nines is 3 seconds.
307:
Availability is usually expressed as a percentage of uptime in a given year. The following table shows the downtime that will be allowed for a particular percentage of availability, presuming that the system is required to operate continuously.
1192:
Sometimes the humorous term "nine fives" (55.5555555%) is used to contrast with "five nines" (99.999%), though this is not an actual goal, but rather a sarcastic reference to something totally failing to meet any reasonable target.
1365:
can be used in systems with limited redundancy to achieve high availability. Maintenance actions occur during brief periods of down-time only after a fault indicator activates. Failure is only significant if this occurs during a
1229:(MTTR). Recovery time could be infinite with certain system designs and failures, i.e. full recovery is impossible. One such example is a fire or flood that destroys a data center and its systems when there is no secondary
299:, and application upgrades, patches, and replacements. For certain systems, scheduled downtime does not matter, for example system downtime at an office building after everybody has gone home for the night.
1334:
is derived from early work by Birman and Joseph in this area. Active redundancy may introduce more complex failure modes into a system, such as continuous system reconfiguration due to faulty voting logic.
1760:
1172:
In general, the number of nines is not often used by a network engineer when modeling and measuring availability because it is hard to apply in formula. More often, the unavailability expressed as a
283:
If users can be warned away from scheduled downtimes, then the distinction is useful. But if the requirement is for true high availability, then downtime is downtime whether or not it is scheduled.
2126:
42:
Modernization has resulted in an increased reliance on these systems. For example, hospitals and data centers require high availability of their systems to perform routine daily activities.
1153:
1590:
1053:
268:
components (or possibly other failed hardware components), an over-temperature related shutdown, logically or physically severed network connections, security breaches, or various
830:
1225:(RTO) is closely related to availability, that is the total time required for a planned outage or the time required to fully recover from an unplanned outage. Another metric is
972:
898:
1311:
High availability requires less human intervention to restore operation in complex systems; the reason for this being that the most common cause for outages is human error.
1206:
processing systems or telephone switches, are often inherently better monitored, at least by the users themselves, than systems which experience periodic lulls in demand.
291:. Systems that exhibit truly continuous availability are comparatively rare and higher priced, and most have carefully implemented specialty designs that eliminate any
1326:. Malfunction of single components is not considered to be a failure unless the resulting performance decline exceeds the specification limits for the entire system.
936:
865:
251:
that is disruptive to system operation and usually cannot be avoided with a currently installed system design. Scheduled downtime events might include patches to
1244:
and other information storage systems faithfully record and report system transactions. Information management often focuses separately on data availability, or
1102:
1079:
791:
739:
are often used interchangeably but do not always refer to the same thing. For example, a system can be "up" with its services not "available" in the case of a
1695:
Smith, Paul; Hutchison, David; Sterbenz, James P.G.; Schöller, Marcus; Fessi, Ali; Karaliopoulos, Merkouris; Lac, Chidung; Plattner, Bernhard (July 3, 2011).
2026:
1764:
1854:
2124:
77:
network, the probable challenges and risks have to be identified and appropriate resilience metrics have to be defined for the service to be protected.
168:
Detection of failures as they occur. If the two principles above are observed, then a user may never see a failure – but the maintenance activity must.
1406:
2194:
2119:
Ulrik Franke, Pontus
Johnson, Johan König, Liv Marcks von Würtemberg: Availability of enterprise IT systems – an expert-based Bayesian model,
2173:
1482:
1918:
1825:
Castet J., Saleh J. Survivability and
Resiliency of Spacecraft and Space-Based Networks: a Framework for Characterization and Analysis",
1350:
events, or system lifetime. Zero downtime involves massive redundancy, which is needed for some types of aircraft and for most kinds of
1952:
1385:
A survey among academic availability experts in 2010 ranked reasons for unavailability of enterprise IT systems. All reasons refer to
1453:, unavailable systems were estimated to have cost American businesses $ 4.54 billion in 1996, due to lost productivity and revenues.
1322:
single engine or propeller. A more complex example is multiple redundant power generation facilities within a large system involving
158:. This means adding or building redundancy into the system so that failure of a component does not mean failure of the entire system.
1937:
989:
1827:
American
Institute of Aeronautics and Astronautics, AIAA Technical Report 2008-7707. Conference on Network Protocols (ICNP 2006)
165:, the crossover point itself tends to become a single point of failure. Reliable systems must provide for reliable crossover.
1633:
1284:
1735:
1004:) 99.999% of the time would have 5 nines reliability, or class five. In particular, the term is used in connection with
204:
1252:
with various failure events. Some users can tolerate application service interruptions but cannot tolerate data loss.
2151:
1993:
1898:
1788:"A Benders Decomposition Approach for Resilient Placement of Virtual Process Control Functions in Mobile Edge Clouds"
1643:
1538:
1110:
230:
212:
2240:
1557:
1477:
1851:
2225:
208:
1295:
Adding more components to an overall system design can undermine efforts to achieve high availability because
1890:
1019:
2235:
1612:
2080:
1977:
35:) is a characteristic of a system that aims to ensure an agreed level of operational performance, usually
1761:"The CERCES project - Center for Resilient Critical Infrastructures at KTH Royal Institute of Technology"
1323:
1159:
796:
2191:
941:
1339:
1301:
1276:
1210:
870:
992:
or "class of nines" in the digits. For example, electricity that is delivered without interruptions (
2068:
but it seems to me we are moving closer to 9-5s (55.5555555%) in network reliability rather than 5-9s
1472:
1355:
2230:
1351:
1314:
1245:
193:
162:
155:
1185:
this is why an "unavailability" rather than availability metric is used in hard disk or data link
1839:
1396:
1373:
1256:
1222:
1009:
309:
292:
288:
197:
147:
137:
2171:
997:
89:
2055:
1280:
1226:
906:
2245:
1956:
1497:
838:
744:
269:
113:
1852:
Introduction to the new mainframe: Large scale commercial computing
Chapter 5 Availability
8:
2204:
1450:
1343:
1807:
1716:
1420:
1087:
1064:
776:
244:
103:
73:
743:. Or a system undergoing software maintenance can be "available" to be worked on by a
2147:
1894:
1659:
1639:
1237:
1230:
108:
72:, the ability to "provide and maintain an acceptable level of service in the face of
1811:
1720:
2104:
1799:
1708:
1367:
1272:
277:
66:
2210:
2198:
2177:
2022:
1858:
1467:
1362:
1180:
per year is quoted. Availability specified as a number of nines is often seen in
252:
243:
A distinction can be made between scheduled and unscheduled downtime. Typically,
1824:
988:
Percentages of a particular order of magnitude are sometimes referred to by the
2108:
2027:"After 35 years of technology crusades, Bob Metcalfe rides off into the sunset"
1787:
1696:
1487:
1296:
1268:
1259:("SLA") formalizes an organization's availability objectives and requirements.
1186:
1082:
740:
143:
2010:
leading to crashes and uptime numbers closer to nine fives than to five nines.
1870:
1803:
1712:
2219:
2130:
1166:
1001:
983:
128:
are interchangeably used according to the specific context of a given study.
125:
95:
20:
1932:
Murphy, Niall
Richard; Beyer, Betsy; Petoff, Jennifer; Jones, Chris (2016).
1915:
2121:
Proc. Fourth
International Workshop on Software Quality and Maintainability
1462:
993:
736:
43:
1677:
1492:
1338:
Zero downtime system design means that modeling and simulation indicates
1173:
639:
576:
248:
773:
Another memory trick to calculate the allowed downtime duration for an "
1617:
1005:
702:
296:
273:
260:
Examples of unscheduled downtime events include power outages, failed
1249:
1181:
1528:
days). For consistency, all times are rounded to two decimal digits.
182:
1520:
days per year; respectively, a quarter is a ÂĽ of that value (i.e.,
1305:
1241:
1177:
748:
53:
1347:
1331:
256:
1934:
Site
Reliability Engineering: How Google Runs Production Systems
1221:
Recovery time (or estimated time of repair (ETR), also known as
732:
36:
1694:
1330:
used with complex computing systems that are linked. Internet
19:"Always-on" redirects here. For the software restriction, see
1992:
Newman, David; Snyder, Joel; Thayer, Rodney (June 24, 2012).
1736:"operational resilience | telcos | accesstel | risk | crisis"
1631:
2146:(Second ed.). Indianapolis, IN: John Wiley & Sons.
1267:
High availability is one of the primary requirements of the
57:
is used to refer to periods when a system is unavailable.
1389:
in each of the following areas (in order of importance):
265:
261:
1441:
A book on the factors themselves was published in 2003.
1169:
is sometimes used to describe the purity of substances.
1931:
793:-nines" availability percentage is to use the formula
295:
and allow online hardware, network, operating system,
100:
maintaining service of communication services such as
1560:
1279:. If the controlling system becomes unavailable, the
1113:
1090:
1067:
1022:
944:
909:
873:
841:
799:
779:
172:
1635:
1550:"Twice as much" on a logarithmic scale, meaning two
1792:
IEEE Transactions on
Network and Service Management
1991:
1584:
1147:
1096:
1073:
1047:
966:
930:
892:
859:
835:For example, 90% ("one nine") yields the exponent
824:
785:
1342:significantly exceeds the period of time between
1196:
768:
2217:
903:Also, 99.999% ("five nines") gives the exponent
1829:, Santa Barbara, California, USA, November 2006
1248:, in order to determine acceptable (or actual)
1148:{\displaystyle c:=\lfloor -\log _{10}x\rfloor }
2042:and five nines (not nine fives) of reliability
1786:Zhao, Peiyue; Dán, György (December 3, 2018).
2205:Lecture notes on Embedded Systems Engineering
1524:days), and a month is a twelfth of it (i.e.,
1380:
747:, but its services do not appear "up" to the
16:Systems with high up-time, a.k.a. "always on"
1585:{\displaystyle \times 2\times 2<\times 5}
1483:Reliability, availability and serviceability
1142:
1120:
1008:or enterprise computing, often as part of a
2141:
1950:
1697:"Network resilience: a systematic approach"
1539:mathematical coincidences concerning base 2
1262:
1216:
211:. Unsourced material may be challenged and
2073:
1444:
1411:Avoidance of internal application failures
150:which can help achieve high availability.
1871:IBM zEnterprise EC12 Business Value Video
1733:
1358:is an example of a zero downtime system.
302:
231:Learn how and when to remove this message
120:access to applications and data as needed
2021:
1994:"Crying Wolf: False alarms hide attacks"
1414:Avoidance of external services that fail
938:, and therefore the allowed downtime is
867:, and therefore the allowed downtime is
755:
2053:
2000:. Vol. 19, no. 25. p. 60
1785:
2218:
1632:Floyd Piedad, Michael Hawkins (2001).
1610:
1048:{\displaystyle \log _{10}2\approx 0.3}
2192:Lecture Notes on Enterprise Computing
1393:Monitoring of the relevant components
1842:M. Nesterenko, Kent State University
1285:ASW Continuous Trail Unmanned Vessel
977:
209:adding citations to reliable sources
176:
1016:5 is 0.3 nines (per below formula:
825:{\displaystyle 8.64\times 10^{4-n}}
760:A simple mnemonic rule states that
65:High availability is a property of
39:, for a higher than normal period.
13:
1541:for details on this approximation.
967:{\displaystyle 8.64\times 10^{-1}}
173:Scheduled and unscheduled downtime
14:
2257:
2185:
2142:Marcus, Evan; Stern, Hal (2003).
2054:Pilgrim, Jim (October 20, 2010).
1678:"Webarchiv ETHZ / Webarchive ETH"
893:{\displaystyle 8.64\times 10^{3}}
2144:Blueprints for high availability
1889:. Pergamon Press. 1981. p.
1290:
709:99.9999999999% ("twelve nines")
181:
2160:
2135:
2113:
2098:
2047:
2015:
1985:
1971:
1944:
1925:
1907:
1879:
1864:
1845:
1833:
1544:
1531:
1510:
1478:Overall equipment effectiveness
1437:Storage architecture redundancy
686:99.999999999% ("eleven nines")
2168:Improving systems availability
1951:Josh Deprez (April 23, 2016).
1818:
1779:
1753:
1727:
1688:
1670:
1660:"Definitions - ResiliNetsWiki"
1652:
1625:
1611:Robert, Sheldon (April 2024).
1604:
1240:, that is the degree to which
1197:Measurement and interpretation
142:There are three principles of
1:
2170:, IBM Global Services, 1998,
1598:
315:
131:
60:
1701:IEEE Communications Magazine
1425:Technical solution of backup
540:99.995% ("four nines five")
500:99.95% ("three nines five")
335:Downtime per day (24 hours)
7:
2081:"What is network downtime?"
1456:
1387:not following best practice
1324:electric power transmission
1277:autonomous maritime vessels
1236:Another related concept is
1160:Floor and ceiling functions
666:99.99999999% ("ten nines")
646:99.9999999% ("nine nines")
623:99.999999% ("eight nines")
10:
2262:
2197:November 16, 2013, at the
1734:accesstel (June 9, 2022).
1428:Process solution of backup
1381:Reasons for unavailability
1340:mean time between failures
1211:mean time between failures
981:
708:
685:
665:
645:
622:
603:99.99999% ("seven nines")
602:
582:
559:
539:
519:
499:
479:
460:99.8% ("two nines eight")
459:
439:
419:
399:
379:
359:
339:
135:
18:
1887:Precious metals, Volume 4
1804:10.1109/TNSM.2018.2873178
1713:10.1109/MCOM.2011.5936160
1473:High-availability cluster
1434:Infrastructure redundancy
1356:Global Positioning System
1352:communications satellites
1209:An alternative metric is
440:99.5% ("two nines five")
1814:– via IEEE Xplore.
1723:– via IEEE Xplore.
1613:"high availability (HA)"
1503:
1263:Military control systems
1246:Recovery Point Objective
1217:Closely related concepts
310:Service level agreements
156:single points of failure
84:These services include:
2241:Reliability engineering
2211:Uptime Calculator (SLA)
1445:Costs of unavailability
1374:Modeling and simulation
1287:(ACTUV) would be lost.
1257:service level agreement
1223:recovery time objective
1010:service-level agreement
583:99.9999% ("six nines")
560:99.999% ("five nines")
400:98% ("one nine eight")
380:97% ("one nine seven")
293:single point of failure
289:continuous availability
161:Reliable crossover. In
148:reliability engineering
138:Design for availability
2201:University of TĂĽbingen
2176:April 1, 2011, at the
1857:March 4, 2016, at the
1586:
1449:In a 1998 report from
1149:
1098:
1075:
1049:
968:
932:
931:{\displaystyle 4-5=-1}
894:
861:
826:
787:
752:be used synonymously.
520:99.99% ("four nines")
480:99.9% ("three nines")
360:95% ("one nine five")
303:Percentage calculation
90:distributed processing
51:. Generally, the term
2226:System administration
2207:by Prof. Phil Koopman
2166:IBM Global Services,
2123:(WSQM 2010), Madrid,
1980:The myth of the nines
1587:
1281:Ground Combat Vehicle
1227:mean time to recovery
1176:(like 0.00001), or a
1150:
1099:
1076:
1058:A formulation of the
1050:
969:
933:
895:
862:
860:{\displaystyle 4-1=3}
827:
788:
756:Five-by-five mnemonic
326:Downtime per quarter
136:Further information:
1959:on September 4, 2016
1558:
1498:Ubiquitous computing
1417:Physical environment
1111:
1088:
1081:based on a system's
1065:
1020:
942:
907:
871:
839:
797:
777:
769:"Powers of 10" trick
745:system administrator
689:315.58 microseconds
675:262.80 microseconds
672:788.40 microseconds
658:604.80 microseconds
626:315.58 milliseconds
612:262.98 milliseconds
595:604.80 milliseconds
320:Availability %
205:improve this section
114:online collaboration
2236:Applied probability
2129:August 4, 2012, at
1767:on October 19, 2018
1451:IBM Global Services
1344:planned maintenance
1167:similar measurement
721:604.81 nanoseconds
712:31.56 microseconds
695:26.28 microseconds
692:78.84 microseconds
678:60.48 microseconds
661:86.40 microseconds
649:31.56 milliseconds
632:26.30 milliseconds
629:78.89 milliseconds
615:60.48 milliseconds
598:86.40 milliseconds
329:Downtime per month
1582:
1421:Network redundancy
1145:
1094:
1071:
1045:
964:
928:
890:
857:
822:
783:
724:86.40 nanoseconds
718:2.63 microseconds
715:7.88 microseconds
698:6.05 microseconds
681:8.64 microseconds
669:3.16 milliseconds
655:2.63 milliseconds
652:7.89 milliseconds
635:6.05 milliseconds
618:8.64 milliseconds
420:99% ("two nines")
332:Downtime per week
323:Downtime per year
245:scheduled downtime
104:video conferencing
2058:. Clearfield, Inc
2056:"Goodbye Five 9s"
2025:(April 2, 2001).
1682:webarchiv.ethz.ch
1638:. Prentice Hall.
1431:Physical location
1273:unmanned vehicles
1238:data availability
1231:disaster recovery
1097:{\displaystyle x}
1074:{\displaystyle c}
974:seconds per day.
900:seconds per day.
832:seconds per day.
786:{\displaystyle n}
728:
727:
340:90% ("one nine")
241:
240:
233:
163:redundant systems
109:instant messaging
29:High availability
2253:
2180:
2164:
2158:
2157:
2139:
2133:
2117:
2111:
2102:
2096:
2095:
2093:
2091:
2077:
2071:
2070:
2065:
2063:
2051:
2045:
2044:
2039:
2037:
2019:
2013:
2012:
2007:
2005:
1989:
1983:
1978:Evan L. Marcus,
1975:
1969:
1968:
1966:
1964:
1955:. Archived from
1953:"Nines of Nines"
1948:
1942:
1941:
1929:
1923:
1922:
1917:. 1998. p.
1911:
1905:
1904:
1883:
1877:
1868:
1862:
1849:
1843:
1837:
1831:
1822:
1816:
1815:
1798:(4): 1460–1472.
1783:
1777:
1776:
1774:
1772:
1763:. Archived from
1757:
1751:
1750:
1748:
1746:
1731:
1725:
1724:
1692:
1686:
1685:
1674:
1668:
1667:
1656:
1650:
1649:
1629:
1623:
1622:
1608:
1592:
1591:
1589:
1588:
1583:
1548:
1542:
1535:
1529:
1527:
1523:
1519:
1514:
1407:network failures
1368:mission critical
1154:
1152:
1151:
1146:
1135:
1134:
1103:
1101:
1100:
1095:
1080:
1078:
1077:
1072:
1054:
1052:
1051:
1046:
1032:
1031:
973:
971:
970:
965:
963:
962:
937:
935:
934:
929:
899:
897:
896:
891:
889:
888:
866:
864:
863:
858:
831:
829:
828:
823:
821:
820:
792:
790:
789:
784:
317:
316:
278:operating system
236:
229:
225:
222:
216:
185:
177:
2261:
2260:
2256:
2255:
2254:
2252:
2251:
2250:
2231:Quality control
2216:
2215:
2199:Wayback Machine
2188:
2183:
2178:Wayback Machine
2165:
2161:
2154:
2140:
2136:
2118:
2114:
2103:
2099:
2089:
2087:
2079:
2078:
2074:
2061:
2059:
2052:
2048:
2035:
2033:
2020:
2016:
2003:
2001:
1990:
1986:
1976:
1972:
1962:
1960:
1949:
1945:
1930:
1926:
1913:
1912:
1908:
1901:
1885:
1884:
1880:
1869:
1865:
1859:Wayback Machine
1850:
1846:
1838:
1834:
1823:
1819:
1784:
1780:
1770:
1768:
1759:
1758:
1754:
1744:
1742:
1732:
1728:
1693:
1689:
1676:
1675:
1671:
1658:
1657:
1653:
1646:
1630:
1626:
1609:
1605:
1601:
1596:
1595:
1559:
1556:
1555:
1549:
1545:
1536:
1532:
1525:
1521:
1517:
1515:
1511:
1506:
1468:Fault tolerance
1459:
1447:
1399:and procurement
1383:
1363:instrumentation
1297:complex systems
1293:
1269:control systems
1265:
1219:
1199:
1187:bit error rates
1130:
1126:
1112:
1109:
1108:
1089:
1086:
1085:
1066:
1063:
1062:
1027:
1023:
1021:
1018:
1017:
990:number of nines
986:
980:
955:
951:
943:
940:
939:
908:
905:
904:
884:
880:
872:
869:
868:
840:
837:
836:
810:
806:
798:
795:
794:
778:
775:
774:
771:
758:
729:
305:
255:that require a
253:system software
247:is a result of
237:
226:
220:
217:
202:
186:
175:
154:Elimination of
140:
134:
124:Resilience and
96:network storage
63:
24:
17:
12:
11:
5:
2259:
2249:
2248:
2243:
2238:
2233:
2228:
2214:
2213:
2208:
2202:
2187:
2186:External links
2184:
2182:
2181:
2159:
2152:
2134:
2112:
2097:
2072:
2046:
2014:
1984:
1970:
1943:
1924:
1906:
1899:
1878:
1863:
1844:
1832:
1817:
1778:
1752:
1726:
1687:
1669:
1664:resilinets.org
1651:
1644:
1624:
1602:
1600:
1597:
1594:
1593:
1581:
1578:
1575:
1572:
1569:
1566:
1563:
1543:
1530:
1508:
1507:
1505:
1502:
1501:
1500:
1495:
1490:
1488:Responsiveness
1485:
1480:
1475:
1470:
1465:
1458:
1455:
1446:
1443:
1439:
1438:
1435:
1432:
1429:
1426:
1423:
1418:
1415:
1412:
1409:
1403:
1400:
1394:
1382:
1379:
1302:load balancing
1292:
1289:
1264:
1261:
1218:
1215:
1198:
1195:
1156:
1155:
1144:
1141:
1138:
1133:
1129:
1125:
1122:
1119:
1116:
1093:
1083:unavailability
1070:
1044:
1041:
1038:
1035:
1030:
1026:
982:Main article:
979:
976:
961:
958:
954:
950:
947:
927:
924:
921:
918:
915:
912:
887:
883:
879:
876:
856:
853:
850:
847:
844:
819:
816:
813:
809:
805:
802:
782:
770:
767:
757:
754:
741:network outage
726:
725:
722:
719:
716:
713:
710:
706:
705:
699:
696:
693:
690:
687:
683:
682:
679:
676:
673:
670:
667:
663:
662:
659:
656:
653:
650:
647:
643:
642:
636:
633:
630:
627:
624:
620:
619:
616:
613:
610:
607:
604:
600:
599:
596:
593:
590:
587:
586:31.56 seconds
584:
580:
579:
573:
570:
569:26.30 seconds
567:
564:
561:
557:
556:
553:
552:30.24 seconds
550:
547:
544:
543:26.30 minutes
541:
537:
536:
533:
530:
527:
526:13.15 minutes
524:
523:52.60 minutes
521:
517:
516:
515:43.20 seconds
513:
510:
509:21.92 minutes
507:
504:
501:
497:
496:
493:
492:10.08 minutes
490:
489:43.83 minutes
487:
484:
481:
477:
476:
473:
472:20.16 minutes
470:
469:87.66 minutes
467:
464:
461:
457:
456:
453:
452:50.40 minutes
450:
447:
444:
441:
437:
436:
435:14.40 minutes
433:
430:
427:
424:
421:
417:
416:
415:28.80 minutes
413:
410:
407:
404:
401:
397:
396:
395:43.20 minutes
393:
390:
387:
384:
381:
377:
376:
373:
370:
367:
364:
361:
357:
356:
353:
350:
347:
344:
341:
337:
336:
333:
330:
327:
324:
321:
304:
301:
239:
238:
189:
187:
180:
174:
171:
170:
169:
166:
159:
144:systems design
133:
130:
122:
121:
118:
117:
116:
111:
106:
98:
92:
62:
59:
15:
9:
6:
4:
3:
2:
2258:
2247:
2244:
2242:
2239:
2237:
2234:
2232:
2229:
2227:
2224:
2223:
2221:
2212:
2209:
2206:
2203:
2200:
2196:
2193:
2190:
2189:
2179:
2175:
2172:
2169:
2163:
2155:
2153:0-471-43026-9
2149:
2145:
2138:
2132:
2131:archive.today
2128:
2125:
2122:
2116:
2110:
2106:
2101:
2086:
2082:
2076:
2069:
2057:
2050:
2043:
2032:
2028:
2024:
2023:Metcalfe, Bob
2018:
2011:
1999:
1998:Network World
1995:
1988:
1982:
1981:
1974:
1958:
1954:
1947:
1939:
1935:
1928:
1920:
1916:
1910:
1902:
1900:9780080253695
1896:
1892:
1888:
1882:
1876:
1872:
1867:
1860:
1856:
1853:
1848:
1841:
1840:Lecture Notes
1836:
1830:
1828:
1821:
1813:
1809:
1805:
1801:
1797:
1793:
1789:
1782:
1766:
1762:
1756:
1741:
1737:
1730:
1722:
1718:
1714:
1710:
1706:
1702:
1698:
1691:
1683:
1679:
1673:
1665:
1661:
1655:
1647:
1645:9780130962881
1641:
1637:
1636:
1628:
1620:
1619:
1614:
1607:
1603:
1579:
1576:
1573:
1570:
1567:
1564:
1561:
1553:
1547:
1540:
1534:
1513:
1509:
1499:
1496:
1494:
1491:
1489:
1486:
1484:
1481:
1479:
1476:
1474:
1471:
1469:
1466:
1464:
1461:
1460:
1454:
1452:
1442:
1436:
1433:
1430:
1427:
1424:
1422:
1419:
1416:
1413:
1410:
1408:
1405:Avoidance of
1404:
1401:
1398:
1395:
1392:
1391:
1390:
1388:
1378:
1375:
1371:
1369:
1364:
1359:
1357:
1353:
1349:
1345:
1341:
1336:
1333:
1327:
1325:
1319:
1316:
1312:
1309:
1307:
1303:
1298:
1291:System design
1288:
1286:
1282:
1278:
1274:
1270:
1260:
1258:
1253:
1251:
1247:
1243:
1239:
1234:
1233:data center.
1232:
1228:
1224:
1214:
1212:
1207:
1203:
1194:
1190:
1188:
1183:
1179:
1175:
1170:
1168:
1163:
1161:
1139:
1136:
1131:
1127:
1123:
1117:
1114:
1107:
1106:
1105:
1091:
1084:
1068:
1061:
1056:
1042:
1039:
1036:
1033:
1028:
1024:
1013:
1011:
1007:
1003:
999:
995:
991:
985:
984:Nine (purity)
975:
959:
956:
952:
948:
945:
925:
922:
919:
916:
913:
910:
901:
885:
881:
877:
874:
854:
851:
848:
845:
842:
833:
817:
814:
811:
807:
803:
800:
780:
766:
763:
753:
750:
746:
742:
738:
734:
723:
720:
717:
714:
711:
707:
704:
700:
697:
694:
691:
688:
684:
680:
677:
674:
671:
668:
664:
660:
657:
654:
651:
648:
644:
641:
637:
634:
631:
628:
625:
621:
617:
614:
611:
609:0.79 seconds
608:
606:3.16 seconds
605:
601:
597:
594:
592:2.63 seconds
591:
589:7.89 seconds
588:
585:
581:
578:
574:
572:6.05 seconds
571:
568:
566:1.31 minutes
565:
563:5.26 minutes
562:
558:
555:4.32 seconds
554:
551:
549:2.19 minutes
548:
546:6.57 minutes
545:
542:
538:
535:8.64 seconds
534:
532:1.01 minutes
531:
529:4.38 minutes
528:
525:
522:
518:
514:
512:5.04 minutes
511:
508:
506:65.7 minutes
505:
502:
498:
495:1.44 minutes
494:
491:
488:
485:
482:
478:
475:2.88 minutes
474:
471:
468:
465:
462:
458:
455:7.20 minutes
454:
451:
448:
445:
442:
438:
434:
431:
428:
425:
422:
418:
414:
411:
408:
405:
402:
398:
394:
391:
388:
385:
382:
378:
374:
371:
368:
365:
362:
358:
354:
351:
348:
345:
342:
338:
334:
331:
328:
325:
322:
319:
318:
314:
311:
300:
298:
294:
290:
284:
281:
279:
275:
271:
267:
263:
258:
254:
250:
246:
235:
232:
224:
214:
210:
206:
200:
199:
195:
190:This section
188:
184:
179:
178:
167:
164:
160:
157:
153:
152:
151:
149:
145:
139:
129:
127:
126:survivability
119:
115:
112:
110:
107:
105:
102:
101:
99:
97:
93:
91:
87:
86:
85:
82:
78:
75:
71:
68:
58:
56:
55:
50:
45:
40:
38:
34:
30:
26:
22:
21:Always-on DRM
2167:
2162:
2143:
2137:
2120:
2115:
2100:
2090:December 27,
2088:. Retrieved
2084:
2075:
2067:
2060:. Retrieved
2049:
2041:
2034:. Retrieved
2030:
2017:
2009:
2002:. Retrieved
1997:
1987:
1979:
1973:
1961:. Retrieved
1957:the original
1946:
1933:
1927:
1914:
1909:
1886:
1881:
1874:
1866:
1847:
1835:
1826:
1820:
1795:
1791:
1781:
1769:. Retrieved
1765:the original
1755:
1743:. Retrieved
1739:
1729:
1707:(7): 88–97.
1704:
1700:
1690:
1681:
1672:
1663:
1654:
1634:
1627:
1616:
1606:
1551:
1546:
1533:
1512:
1463:Availability
1448:
1440:
1397:Requirements
1386:
1384:
1372:
1360:
1337:
1328:
1320:
1313:
1310:
1294:
1266:
1254:
1235:
1220:
1208:
1204:
1200:
1191:
1171:
1164:
1157:
1059:
1057:
1014:
987:
902:
834:
772:
761:
759:
737:availability
730:
640:microseconds
577:milliseconds
463:17.53 hours
446:10.98 hours
409:14.61 hours
406:43.86 hours
389:21.92 hours
369:36.53 hours
352:16.80 hours
349:73.05 hours
306:
285:
282:
242:
227:
218:
203:Please help
191:
141:
123:
83:
79:
69:
64:
52:
48:
44:Availability
41:
32:
28:
27:
25:
2246:Measurement
1875:youtube.com
1493:Scalability
1174:probability
1060:class of 9s
703:nanoseconds
503:4.38 hours
486:2.19 hours
483:8.77 hours
466:4.38 hours
449:3.65 hours
432:1.68 hours
429:7.31 hours
426:21.9 hours
412:3.36 hours
392:5.04 hours
383:10.96 days
375:1.20 hours
372:8.40 hours
363:18.26 days
355:2.40 hours
343:36.53 days
270:application
249:maintenance
94:supporting
88:supporting
49:unavailable
2220:Categories
2085:Networking
1936:. p.
1861:IBM (2006)
1771:August 26,
1618:Techtarget
1599:References
1402:Operations
1315:Redundancy
1006:mainframes
731:The terms
443:1.83 days
423:3.65 days
403:7.31 days
386:2.74 days
366:4.56 days
346:9.13 days
297:middleware
280:failures.
274:middleware
132:Principles
70:resilience
61:Resilience
2062:March 15,
2036:March 15,
2004:March 15,
1740:accesstel
1577:×
1568:×
1562:×
1283:(GCV) or
1250:data loss
1242:databases
1182:marketing
1143:⌋
1137:
1124:−
1121:⌊
1104:would be
1040:≈
1034:
998:brownouts
994:blackouts
957:−
949:×
923:−
914:−
878:×
846:−
815:−
804:×
221:June 2008
192:does not
2195:Archived
2174:Archived
2127:Archived
1891:page 262
1855:Archived
1812:56594760
1721:10246912
1457:See also
1370:period.
1306:failover
1213:(MTBF).
1178:downtime
749:end user
54:downtime
2031:ITworld
1963:May 31,
1552:factors
1526:30.4375
1522:91.3125
1348:upgrade
1332:routing
978:"Nines"
762:5 nines
701:864.00
638:864.00
575:864.00
213:removed
198:sources
67:network
2150:
2107:
1897:
1810:
1745:May 8,
1719:
1642:
1554:of 2:
1518:365.25
1516:Using
1361:Fault
1002:surges
733:uptime
276:, and
257:reboot
74:faults
37:uptime
1808:S2CID
1717:S2CID
1504:Notes
1158:(cf.
2148:ISBN
2092:2023
2064:2019
2038:2019
2006:2019
1965:2016
1895:ISBN
1773:2023
1747:2023
1640:ISBN
1574:<
1537:See
1304:and
1275:and
946:8.64
875:8.64
801:8.64
735:and
196:any
194:cite
2109:992
2105:RFC
1919:387
1873:at
1800:doi
1709:doi
1308:).
1271:in
1162:).
1128:log
1043:0.3
1025:log
1000:or
266:RAM
264:or
262:CPU
207:by
146:in
2222::
2083:.
2066:.
2040:.
2029:.
2008:.
1996:.
1938:38
1893:.
1806:.
1796:15
1794:.
1790:.
1738:.
1715:.
1705:49
1703:.
1699:.
1680:.
1662:.
1615:.
1354:.
1346:,
1255:A
1189:.
1165:A
1132:10
1118::=
1029:10
1012:.
996:,
953:10
882:10
808:10
272:,
33:HA
2156:.
2094:.
1967:.
1940:.
1921:.
1903:.
1802::
1775:.
1749:.
1711::
1684:.
1666:.
1648:.
1621:.
1580:5
1571:2
1565:2
1140:x
1115:c
1092:x
1069:c
1037:2
960:1
926:1
920:=
917:5
911:4
886:3
855:3
852:=
849:1
843:4
818:n
812:4
781:n
234:)
228:(
223:)
219:(
215:.
201:.
31:(
23:.
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.
