Hi Thomas, I would have to say that doesn't look right.
Firstly, the chassis figure should have been the same as the per unit figure because there is only one. Then you have to consider whether you are calculating the chances of a fault of any type (sum of parts) or the chance of a service affecting incident. This will affect how you apply the summing or paralleling (for want of a better word). For instance, there are 2 power supplies. Does this make the system more reliable or less. In terms of the chance of a fault occurring, less reliable because there are more components (summing effect). However, a single power supply fault does not stop the unit operating so the parallel effect applies and the system reliability is enhanced. Similarly, the GBICs. There are a lot of them but if they are in a LAG then those do not individually affect service (maybe a bit of bandwidth degradation). It can be a complex thing and you need to be mindful of what you want to know and "sanity check" the numbers. I would suggest that the Chassis is operational reliability for instance is something like Chas+(PS*PS). Normalise your units on per year probability as a unit and not a percentage while calculating so you don't have a 100 fold slipup in figure unit inconsistency. Bruce O'Donnell Senior Network Specialist CCIE NEC Australia Pty Ltd 26 Rodborough Road Frenchs Forest NSW 2086 P: + 61 2 9930 2595 F: + 61 2 9930 2666 M: 0413 181 610 www.nec.com.au CAUTION - This message is intended for the addressee named above. It may contain privileged or confidential information. If you are not the intended recipient of this message your must: - Not use, copy, distribute or disclose it to anyone other than the addressees; - Notify the sender via return email; and - Delete the message (and any related attachments) from your computer immediately. Internet emails are not necessarily secure. NEC Australia ABN 86 001 217 527, do not accept responsibility for changes made to this message after it was sent. Unless otherwise stated, views expressed within this email are the author's own and do not represent those of NEC. -----Original Message----- From: [email protected] [mailto:[email protected]] Sent: Friday, 5 June 2009 4:50 PM To: Enterasys Customer Mailing List Subject: RE: [enterasys] Reliability calculator... Hi William, That sound pretty good :) Over all the years we had only one single chassis fail due to some hardware real time clock error. But I totally agree, those boxes are extremely solid. Why I have been asking about such a tool is that I have been ask to "show" the network availability in numbers, or how reliable the components are. I know that these are just theory, but I think also interesting. I just did some work in excel to calculate it (hopefully I have done it right ;) Maybe someone will find it useful. I also used some sample figures for employees and money to give a rough sum about the costs of a failure. Row E means if you have two physical parts that make up one "logical", like having 36 mgbics, and always two of them are grouped via LAG. So you have 36 hardware pieces but only 18 "virtual" components... Bye Tom -----Original Message----- From: William Olive [mailto:[email protected]] Sent: Friday, June 05, 2009 12:08 AM To: Enterasys Customer Mailing List Subject: Re: [enterasys] Reliability calculator... Tom I can't answer your question, except to say that we run a large number of DFEs and, alone in my experience of network equipment, I have never seen one fail. Like the Cabletron box they evolved from, they are bomb proof. Billo Data Communications Co-Ordinator Information Technology & Telecommunications Hunter New England Health Service ph 0249 213804 fax 0249 213038 [email protected] >>> <[email protected]> Thursday, 4 June 2009 >>> Hi, Just some non-technical question... maybe someone can help me out with this. I am looking for a tool to calculate the reliability of our network components. I found a lot of online calculators that considers the MTBF, MTBR, etc. for just one piece, but I want to do this for multiple pieces and then "sum" it up. For example I would like to know the reliability of a N3, consisting of many different components like the two power supplies, chassis, blades, MGBIC ports, etc Any ideas? BR Tom Hofmann -- Thomas Hofmann, System-/Networkadministrator, IT EB - Discover the Experience Visitors: Am Wolfsmantel 46, 91058 Erlangen, Germany Phone: +49 (9131) 7701 6969, mailto:[email protected] <mailto:[email protected]> Fax: +49 (9131) 7701 6333, http://www.elektrobit.com <http://www.elektrobit.com/> PGP-Key: http://keyserver.elektrobit.com <http://keyserver.elektrobit.com/> Elektrobit Automotive GmbH, Am Wolfsmantel 46, 91058 Erlangen, Germany Managing Director Ottoel Register Court��rth HRB 4886 ---------------------------------------------------------------- Please note: This e-mail may contain confidential information intended solely for the addressee. If you have received this e-mail in error, please do not disclose it to anyone, notify the sender promptly, and delete the message from your system. Thank you. --- To unsubscribe from enterasys, send email to [email protected] with the body: unsubscribe enterasys [email protected] --- To unsubscribe from enterasys, send email to [email protected] with the body: unsubscribe enterasys [email protected] ---------------------------------------------------------------- Please note: This e-mail may contain confidential information intended solely for the addressee. If you have received this e-mail in error, please do not disclose it to anyone, notify the sender promptly, and delete the message from your system. Thank you. --- To unsubscribe from enterasys, send email to [email protected] with the body: unsubscribe enterasys [email protected] --- To unsubscribe from enterasys, send email to [email protected] with the body: unsubscribe enterasys [email protected]
