Re: How long does an I/O take (very roughly)?
Peter, ROTFL a cache miss will never take just 1ms. Seek and latency are still alive and well, so you are looking at 6-9ms for a read cache miss in a relatively idle system. For read cache hit a of a 4KB block will be around 0.2-0.3 ms, again in an idle system. For a write hit the copy products in use can alter the response time significantly, from a few 100ths of ms for in system copy to several ms for synchronous remote copy. Again this assumes activity on a relatively idle machine. If you are using internal SATA drives, FlashDrives, or virtualized storage there will be an even greater range of rough figures to choose from. For the original OP's purposes I would suggest 0.5ms for cache hit and 9ms for cache miss would represent numbers for 4KB blocks in a generic storage system under load. Ron -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Hal Merritt Sent: Wednesday, January 06, 2010 9:14 AM To: IBM-MAIN@bama.ua.edu Subject: Re: [IBM-MAIN] How long does an I/O take (very roughly)? Ignoring a lot and oversimplifying, I believe that modern DASD should routinely deliver 0.1ms (hit) and 1ms (miss). -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Hunkeler Peter (KIUP 4) Sent: Wednesday, January 06, 2010 10:08 AM To: IBM-MAIN@bama.ua.edu Subject: How long does an I/O take (very roughly)? Hi, I'm about to update a chart I'm using to illustrate the relative speed of various actions like CP cycle, storage access, etc. Does it still hold true that an I/O from controller cache (cache hit) takes about 1ms and and I/O from the platter (cache miss) takes about 5ms. All I want is two very rough figures. I don't want to consider different hardware or attachement types. -- Peter Hunkeler CREDIT SUISSE AG NOTICE: This electronic mail message and any files transmitted with it are intended exclusively for the individual or entity to which it is addressed. The message, together with any attachment, may contain confidential and/or privileged information. Any unauthorized review, use, printing, saving, copying, disclosure or distribution is strictly prohibited. If you have received this message in error, please immediately advise the sender by reply email and delete all copies. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
How long does an I/O take (very roughly)?
Hi, I'm about to update a chart I'm using to illustrate the relative speed of various actions like CP cycle, storage access, etc. Does it still hold true that an I/O from controller cache (cache hit) takes about 1ms and and I/O from the platter (cache miss) takes about 5ms. All I want is two very rough figures. I don't want to consider different hardware or attachement types. -- Peter Hunkeler CREDIT SUISSE AG -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: How long does an I/O take (very roughly)?
With a DS8300 Turbo, my overall average response time is around 1ms. It looks to me like a cache hit is a little less than .5ms, with less than .300ms of connect time. Hunkeler Peter (KIUP 4) peter.hunke...@credit-suisse.com 1/6/2010 11:07 AM Hi, I'm about to update a chart I'm using to illustrate the relative speed of various actions like CP cycle, storage access, etc. Does it still hold true that an I/O from controller cache (cache hit) takes about 1ms and and I/O from the platter (cache miss) takes about 5ms. All I want is two very rough figures. I don't want to consider different hardware or attachement types. -- Peter Hunkeler CREDIT SUISSE AG -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html CONFIDENTIALITY/EMAIL NOTICE: The material in this transmission contains confidential and privileged information intended only for the addressee. If you are not the intended recipient, please be advised that you have received this material in error and that any forwarding, copying, printing, distribution, use or disclosure of the material is strictly prohibited. If you have received this material in error, please (i) do not read it, (ii) reply to the sender that you received the message in error, and (iii) erase or destroy the material. Emails are not secure and can be intercepted, amended, lost or destroyed, or contain viruses. You are deemed to have accepted these risks if you communicate with us by email. Thank you. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: How long does an I/O take (very roughly)?
Ignoring a lot and oversimplifying, I believe that modern DASD should routinely deliver 0.1ms (hit) and 1ms (miss). -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Hunkeler Peter (KIUP 4) Sent: Wednesday, January 06, 2010 10:08 AM To: IBM-MAIN@bama.ua.edu Subject: How long does an I/O take (very roughly)? Hi, I'm about to update a chart I'm using to illustrate the relative speed of various actions like CP cycle, storage access, etc. Does it still hold true that an I/O from controller cache (cache hit) takes about 1ms and and I/O from the platter (cache miss) takes about 5ms. All I want is two very rough figures. I don't want to consider different hardware or attachement types. -- Peter Hunkeler CREDIT SUISSE AG NOTICE: This electronic mail message and any files transmitted with it are intended exclusively for the individual or entity to which it is addressed. The message, together with any attachment, may contain confidential and/or privileged information. Any unauthorized review, use, printing, saving, copying, disclosure or distribution is strictly prohibited. If you have received this message in error, please immediately advise the sender by reply email and delete all copies. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html