× The internal search function is temporarily non-functional. The current search engine is no longer viable and we are researching alternatives.
As a stop gap measure, we are using Google's custom search engine service.
If you know of an easy to use, open source, search engine ... please contact support@midrange.com.



>I'd be grateful for some help in understanding the write efficiency
parameter
>in the Disk Activity section of the Performance Monitor Component Report.

John - 

This might help explain things.... (Taken from the BEST/1 manual)


Considering the High Availability models (9337-xx2, 9337-xx7, 660x-070,
671x-070, or 690x-070), another element of performance in addition to the
proportion of writes, is relative cache efficiency. Cache efficiency
describes the relationship of incoming requests for write I/Os and how many
physical writes this actually generates at an arm. If 1000 writes are
directed to the device, 1000 are written to the cache. If only 500 physical
writes to the arm occur, this would result in a cache efficiency of 50%. A
lower number of physical writes might occur because:
 
   Information is getting updated in the cache due to subsequent writes of
the same information. Since the information is already in the cache, no
physical I/O is involved.
 
   When the cache becomes full, data has to be physically written to disk,
and the related pieces of information located near each other on the disk
are merged into one I/O.
 
 
The relative cache efficiency has a large effect on the RAID (hardware
checksum) activity at the arm, because only the physical writes generate
physical checksum activity. BEST/1 reduces the additional arm I/Os due to
checksum activity based on the current cache efficiency. For example, a
value of 50% would generate only half of the otherwise expected additional
checksum I/O activity.
 
In actuality, various workloads experience cache efficiencies ranging from
about 5 to 95%. Thus, if the 9337-xx2, 9337-xx7, 660x-070, 671x-070, or
690x-070), arms and the workloads being modeled would produce a less
favorable (lower) value than the current value for that controller, actual
performance will be worse than predicted. If the workloads being modeled
would produce a good (higher) value, actual performance will be better than
predicted.
 
Cache efficiency values are set in the following ways:
 
   If the model was created from performance data with 9337-xx2, 9337-xx7,
660x-070, 671x-070, or 690x-070), controllers, the cache efficiency is
determined by the actual data.
 
   If the controller or disk IOA was created in the Work with Disk
Resources display, the cache efficiency is taken from the hardware table
value.
 
   If the controller was copied from another controller, the cache
efficiency value is the same as the controller being copied.
 
   If the disk IOA was copied from another disk IOA, the cache efficiency
value is the same as the disk IOA being copied.
 

Kenneth

--
********************************
         Kenneth  E.  Graap             
    IBM Certified Specialist                 
          AS/400   Professional                   
          System  Administrator                  
 NW Natural - Information Services        
           System Services                         
        503 226 4211 X5537                 
          FAX  503 721 2521                  
      keg@nwnatural.com               
********************************


-----Original Message-----
From: john.lewis-crosby@oxinst.co.uk
[mailto:john.lewis-crosby@oxinst.co.uk]
Sent: Thursday, April 08, 1999 9:14 AM
To: MIDRANGE-L@midrange.com
Subject: Write efficiency
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---


As an Amazon Associate we earn from qualifying purchases.

This thread ...


Follow On AppleNews
Return to Archive home page | Return to MIDRANGE.COM home page

This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].

Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.