Would you let us know where you saw a reference to this rule? This is not something we're familiar with but perhaps if we view it in context we can assist further.
Good point! Mention "Rule of 17" around anyone who has been managing Symms for more than a year or two and they will probably assume a reference to the point I described, but there is more to the world than Symmetrix Management (for most people at least).
Thank you for reminding me that the context is every bit as important as the question itself.
Another possible answer is - Rule 17 is also the name for one of the Federal Rules of Civil Proceedure (FRCP) (http://www.law.cornell.edu/rules/frcp/Rule17.htm) This rule has to do with the capacity of a party involved to sue or be sued. FRCP rules now include digital discovery so the rules are very important when considering GRC issues, information lifecycle management policies, and BC/DR decisions.
What is the context of the question? I think that would help ...
The Rule of 17 is a Symmetrix configuration guideline today on current generations of Symm hardware (V-Max). It goes way back to the older Symm models (from before my time in Storage) where you had to configure directors on the Symm in pairs that were spread over redundant backend power zones and or busses. Basically there were 16 slots for directors and they were always installed in pairs so that the director slot numbers added up to 17. Slot 1 would pair up with slot 16... slot 8 pairs up with slot 9. The configuration ensured that no single component failure in the array would cause data unavailability (assuming you follow HA host configuration guidelines).
With the current line of Symmetrix arrays (V-Max) there is no longer a hard and fast requirement to follow this rule. There is also no technical advantage to following this rule. The only real advantage is that there are a lot of "old timer" engineers and specialists who have this rule deeply ingrained in their minds. If you are in an environment where this makes a difference then it can't hurt to continue following the Rule of 17.
Personally we threw out this rule when we deployed V-Max. There are other configurations that make more sense for us, although we still ensure that HA connections from hosts and RDF links span different engines and IO modules. As long as this is accounted for there is no reason to conform to a specific configuration "rule" any more.
Oh how I wish the reason could have been a little more glamorous and somehow to do with prime numbers.
You know, there are two particular species of Cicada (sometimes colloquially referred to as locusts) called Magicicada septendecim and Magicicada tredecim. These two species often live in the same forest environment, having a life cycle of 17 years and 13 years respectively. For the most part of their lives they live underground, feeding on the sap of tree roots. In their final year they emerge from the ground, having metamorphosed from nymphs into adults. It is one of nature’s wonders, every 17 years Magicicada septendecim takes over the forest in a single night; they do what comes naturally to them - sing, eat, mate, lay eggs, and then they die a few weeks later. It is thought that it may be no coincidence that each species has a life cycle equal to a prime number of years; they will be synchronized to emerge in the same year much less frequently than if their life cycles were composite numbers. In fact, they should only have to share and compete for the forest every 13 x 17 = 221 years. If their life cycles were, for example, 18 years and 12 years, then they would compete much more frequently – every 36, 72, 108, 144, 180, and 216 years (6 times more often).
(There are some other possible reasons as to why these life cycles are a prime number of years).
Hello Allen... well, it is pretty cool. I guess in both cases, the rule of 17 for the Symm and the rule of 17 for the Magicicada septendecim, there is a common factor - survival.
DGraci
408 Posts
0
May 31st, 2011 06:00
Good morning -
Would you let us know where you saw a reference to this rule? This is not something we're familiar with but perhaps if we view it in context we can assist further.
Kind regards,
Debi
Allen Ward
2.1K Posts
0
May 31st, 2011 13:00
Good point! Mention "Rule of 17" around anyone who has been managing Symms for more than a year or two and they will probably assume a reference to the point I described, but there is more to the world than Symmetrix Management (for most people at least).
Thank you for reminding me that the context is every bit as important as the question itself.
Anonymous
5 Practitioner
5 Practitioner
•
274.2K Posts
0
May 31st, 2011 13:00
Hello,
Another possible answer is - Rule 17 is also the name for one of the Federal Rules of Civil Proceedure (FRCP) (http://www.law.cornell.edu/rules/frcp/Rule17.htm) This rule has to do with the capacity of a party involved to sue or be sued. FRCP rules now include digital discovery so the rules are very important when considering GRC issues, information lifecycle management policies, and BC/DR decisions.
What is the context of the question? I think that would help ...
thanks,
Wayne
Allen Ward
2.1K Posts
1
May 31st, 2011 13:00
The Rule of 17 is a Symmetrix configuration guideline today on current generations of Symm hardware (V-Max). It goes way back to the older Symm models (from before my time in Storage) where you had to configure directors on the Symm in pairs that were spread over redundant backend power zones and or busses. Basically there were 16 slots for directors and they were always installed in pairs so that the director slot numbers added up to 17. Slot 1 would pair up with slot 16... slot 8 pairs up with slot 9. The configuration ensured that no single component failure in the array would cause data unavailability (assuming you follow HA host configuration guidelines).
With the current line of Symmetrix arrays (V-Max) there is no longer a hard and fast requirement to follow this rule. There is also no technical advantage to following this rule. The only real advantage is that there are a lot of "old timer" engineers and specialists who have this rule deeply ingrained in their minds. If you are in an environment where this makes a difference then it can't hurt to continue following the Rule of 17.
Personally we threw out this rule when we deployed V-Max. There are other configurations that make more sense for us, although we still ensure that HA connections from hosts and RDF links span different engines and IO modules. As long as this is accounted for there is no reason to conform to a specific configuration "rule" any more.
DGraci
408 Posts
0
May 31st, 2011 13:00
Allen,
Thank you so much for providing this explanation!
Regards,
Debi
vinittyagi_4ba06d
10 Posts
0
June 1st, 2011 04:00
Thanks Allen
Richard_Butler
84 Posts
0
June 1st, 2011 10:00
Oh how I wish the reason could have been a little more glamorous and somehow to do with prime numbers.
You know, there are two particular species of Cicada (sometimes colloquially referred to as locusts) called Magicicada septendecim and Magicicada tredecim. These two species often live in the same forest environment, having a life cycle of 17 years and 13 years respectively. For the most part of their lives they live underground, feeding on the sap of tree roots. In their final year they emerge from the ground, having metamorphosed from nymphs into adults. It is one of nature’s wonders, every 17 years Magicicada septendecim takes over the forest in a single night; they do what comes naturally to them - sing, eat, mate, lay eggs, and then they die a few weeks later. It is thought that it may be no coincidence that each species has a life cycle equal to a prime number of years; they will be synchronized to emerge in the same year much less frequently than if their life cycles were composite numbers. In fact, they should only have to share and compete for the forest every 13 x 17 = 221 years. If their life cycles were, for example, 18 years and 12 years, then they would compete much more frequently – every 36, 72, 108, 144, 180, and 216 years (6 times more often).
(There are some other possible reasons as to why these life cycles are a prime number of years).
Allen Ward
2.1K Posts
0
June 1st, 2011 19:00
Isn't that just a little too coincidental? The critters use the "rule of 17" for lifecycle management...
Richard_Butler
84 Posts
0
June 2nd, 2011 04:00
Hello Allen... well, it is pretty cool. I guess in both cases, the rule of 17 for the Symm and the rule of 17 for the Magicicada septendecim, there is a common factor - survival.