Re: RE: Slightly-OT: Throw HW at a SW/DB problem

From: Niall Litchfield <>
Date: Mon, 27 Jun 2011 18:51:17 +0100
Message-ID: <>

I do trust you. I know this as well. In this case (and many others) the design was evidently not terrible. In fact several weeks of effort improved things by only 20% - that suggests a fundamentally reasonable design to me. You're unlikely to get me of all people arguing for the blind application of hardware, but you won't find me saying "ah its always the app", sometimes it really isn't - otherwise none of you would run UNIX you'd stick with dual processor x86 and Linux/Windows, after all platform choice is irrelevant. Where I do disagree with the article is in the art v science bit - you should be able to predict your improvement.

On 27 Jun 2011 17:33, "Michael Dinh" <> wrote:

 Faster CPU, more cores, more RAM, faster disks, SAME, all those do not help a poor design. Trust me from experience.****

  • **

Is the bottleneck at IO or code or design?****

  • **

Michael Dinh ****

NOTICE OF CONFIDENTIALITY - This material is intended for the use of the individual or entity to which it is addressed, and may contain information that is privileged, confidential and exempt from disclosure under applicable laws. BE FURTHER ADVISED THAT THIS EMAIL MAY CONTAIN PROTECTED HEALTH INFORMATION (PHI). BY ACCEPTING THIS MESSAGE, YOU ACKNOWLEDGE THE FOREGOING, AND AGREE AS FOLLOWS: YOU AGREE TO NOT DISCLOSE TO ANY THIRD PARTY ANY PHI CONTAINED HEREIN, EXCEPT AS EXPRESSLY PERMITTED AND ONLY TO THE EXTENT NECESSARY TO PERFORM YOUR OBLIGATIONS RELATING TO THE RECEIPT OF THIS MESSAGE. If the reader of this email (and attachments) is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. Please notify the sender of the error and delete the e-mail you received. Thank you.****

*From:* []
*On Behalf Of *Niall Litchfield
*Sent:* Monday, June 27, 2011 8:22 AM


Subject: Re: Slightly-OT: Throw HW at a SW/DB problem

  • **

I'm not sure how many people are advocating hardware based tuning - but I do have *some *sympathy with what the guy is saying. We don't get to see what the tuning process was, but a quick triage might go along the lines of ****

"this problem workload currently does 10 million I/Os at 10ms each on average, that takes nearl...

Received on Mon Jun 27 2011 - 12:51:17 CDT

Original text of this message