Re: Performance / Internal Q's -- Online 7.2

This is a multi-part message in MIME format. --------------24681CC654CE Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit jimlenahan@aol.com wrote: > > I have a 4way HP Kclass machine as server connected to a pair of HP > G class machines. On the G's are up to 500 users. The K server is > configured with 4 net VP's with 250 threads each. However, the client > software is going to its knees at times as it waits for polling threads to > become available. Mhy questions are: > > Why is it recommended that the number of SOC VP's not exceed > NUMCPUVP? Increasing the number of VP's seems like it would solve the > problem, but the IFMX docs and classes specifically say not to exceed that > number. I think this recommendation was for CPU VPs, and is a practical limit of one VP per physical CPU. The actual recommendation is the number of CPUs less one. Some platforms can handle more, try it and see. > Currently, the config file has a line for the net VP's with a VP type > of SOC. Can this line be replicated with a VP type of CPU? THis would > give double the number of net VP's (4 each of CPU and SOC), while still > preserving the letter of the law that the number of SOC type VP's not > exceed MAXCPUVP. However, will the engine read BOTH lines, or will the > second effectively replace the first? I am not sure if I am confused or you. If you are referring to the NETTYPEdefinitions, then the only values you can have are NET or CPU, not SOC. SOC would presumably refer to sockets as opposed to TLI (Transport Layer Interface). Your NETTYPE lines should look similar to the following: NETTYPE ipcshm,1,100,CPU NETTYPE soctcp,1,100,NET This example allocates 1 poll thread for SHM (shared memory) access to run on the CPU VP, and 1 poll thread for TCP (TCP/IP using Sockets) access to run on the NETwork (in this case the SOC) VP. Both of these configurations allow for 100 users on each protocol. You can configure poll threads to run on either the CPU VP or the appropriate NETwork VP (SOC, TLI, or SHM). However, only one protocol can run a poll thread on the CPU VP. The protocol that requires best performance should be run on the CPU VP. NOTE: a single thread should be able to handle up to 100 users. You can do the maths for your site. :-) > How much overhead is there in these threads? Does it matter to have > the number of threads per VP set so high? Also, at what point are they > (the threads) recylced? Upon disconnect? I would guess it is an overhead, but I do not know the internals. Remember that if the poll thread is running on the SOC, TLI, or SHM VPs, by specifying the number of poll threads, you are also specifying the number of SOC, TLI, or SHM VPs. If the poll thread will run on the CPU VP, you cannot specify more poll threads than the number of CPU VPs allocated. > A few more related performance / internal questions: > - The CPU's on this server system are not taxed (typically running > at 50-80%). What are the ramifications of increasing MAXCPUVP to be > greater than the actual number of CPU's? Again, IFMX recommends against > it -- but if the code is written right it oughtn't matter . . . Suck it and see. :-) Hope this helps, -- --------------24681CC654CE Content-Type: text/plain; charset=us-ascii; name="ATT_SIG.TXT" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="ATT_SIG.TXT" Mark. +------------------------------------------------------------------+ |Mark D. Stock - The West Solutions Group http://www.west.co.za | | The Informix FAQ is at http://www.iiug.org | |mailto:marks@west.co.za +-----------------------------------------+ |Tel: +27 11 803 2151 |If it doesn't work, force it! | |Fax: +27 11 803 2189 |If it breaks, it needed replacing anyway!| |Cell: +27 83 250 2325 |Well, that's how I code anyway! | +------------------------+-----------------------------------------+ --------------24681CC654CE--