SQL Server 2012 Note Sharing-52: Availability Metrics

In telecommunications and reliability theory, usability refers to:

The degree to which a system, subsystem, or device is in a specified operational or actionable state at the beginning of a task, when the task is used to be unknown, for example, random. Simply put, usability is the percentage of time a system is in a working state. This is usually described as a task-feasible rate. Mathematically speaking, the equivalent of 1 minus the unavailability.

The percentage of total available time for a functional individual in a given time interval.

For example, a unit with a 100-hour availability for a week (168 hours) is 100/168 available. The value of the availability is usually expressed in decimals (such as 0.9998). In high-availability applications, use a metric called several nine, corresponding to the number of 9 after the decimal point. In this system, "Five Nine" is equivalent to 0.99999 (or 99.999%) of the availability.

Example

If we are using a device with an MTBF (mean failure interval) of 81.5 years, MDT (average repair time) is 1 hours:

MTBF in hours = 81.5*365*24=713940

availability= mtbf/(MTBF+MDT) = 713940/713941 =99.999859%

unavailability = 0.000141%

Each year the machine time per device is measured in hours: u=0.01235 hours per year.

==============================================================

The definition in ISO9241/11 is: a product can be a specific user in a particular situation, effective, efficient and satisfied with the extent of achieving a specific goal (the extent to which a product can be used by specified users to AC Hieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use.).

gb/t3187-97 definition of availability: the ability of a product to function in an enforceable state under specified conditions and within a specified time or timeframe, provided the required external resources are guaranteed. It is a comprehensive reflection of the reliability, maintainability and maintenance of the product.

==============================================================

The following is a legend of availability, under different availability criteria, the allowable downtime per year, the allowable downtime per month, and the weekly downtime allowed.

650) this.width=650; "title=" clipboard "style=" Border-top:0px;border-right:0px;background-image:none; border-bottom:0px;padding-top:0px;padding-left:0px;border-left:0px;padding-right:0px, "border=" 0 "alt=" clipboard "Src=" Http://img1.51cto.com/attachment/201407/6/639838_1404611946XiqH.png "height=" 399 "/>

==============================================================

Extended knowledge

RPO (Recovery point Object)

Refers to a past point in time when a disaster or emergency occurs and the data can be restored to a point in time. For example, daily 23:00 for data backup, if there is an outage today, the data can be restored to a point in time (RPO) is yesterday's 23:00.

(compared to RTO, recovery time objective refers to how long it takes to resume running after an outage.) )

A short RPO can lose data less. For example, a five-minute RPO indicates that data must be recovered in five minutes, and an one-hour RPO indicates that the data recovery is a weakness in which the data to be backed up may have been lost within the one-hour period. Conversely, a 0-minute RPO indicates that no data can be lost because your data is backed up, copied, or recorded in a timely manner, preventing any loss of data. Another aspect of RPO to consider is the degree to which data protection is complete and comprehensive. For example, if your RPO is backed up every 8 hours, it means that the data may be lost within 8 hours. Complete and comprehensive data protection focuses on whether your data is 100% protected or that only some of the files and data are protected. For another example, open files may not be fully backed up, unless the data in the cache is stored in the memory on the disk. Another factor to consider is whether the file you are backing up is a particular file in a particular directory or file share, and whether the data is fully backed up. A small rpo means more expense and less data loss, and we have to make a tradeoff between this.

Simply put: The maximum allowable data loss when a failure occurs.

RTO: (Recoverytime Object) refers to the time period between this two-point period, when a disaster occurs, starting with an IT system outage leading to business downtime, to the recovery of IT systems to support all departments, and when business resumes operations.

Simply put: The maximum amount of downtime allowed when a failure occurs, usually expressed as a number, such as 9s.

The higher the goal, the higher the cost.

=================================================================

The myth of the 9 ' s of availability

It is common-organizations to state that they provide a number of 9 's of availability when referring to their environm Ents. The truth is often much different than what's advertised and even then, it's often meant for only operating hours or not Counting planned downtime, which May is not being clearly documented in the SLA. Committing to hours and unplanned outages is acceptable as long as it's supported by what's documented in The SLA.

Note:microsoft recommends that the 9 's of availability is based on agreed upon hours of operation, which should is clear Ly stated in the SLA.

The table on the slide above outlines the 9 's of availability and what actually means to has that level of uptime. Based on the table above, if-organization claims to has 3–9 ' of availability and they is a 24/7 operation, they CA N only has 8.76 hours of downtime per year.

Additional Resources

The table above provides only a brief idea of availability impact and understanding high availability for operations. For more information, refer to the following Microsoft Operations Framework (MOF) Resources:

Microsoft Operations Framework–sla review–

Http://www.microsoft.com/technet/solutionaccelerators/cits/mo/mof/omr/sla.mspx

High availability and the Microsoft Operations framework–

Http://technet.microsoft.com/en-us/library/aa560207.aspx

=================================================================

This article from "Zeng Hung Xin Technical column" blog, declined to reprint!