Re: XPS max I/OMB per CPU second

No hands on with XPS, but if a CPU is sequentially scanning a single table and that single table is on a single disk, 5-10MB/sec is about right for the sequential bandwidth of a SCSI or FCAL disk. Have you tried striping (or using Informix fragmentation on) the big tables you tend to scan to improve parallelism?? The increase in bandwidth should be nearly linear. Assuming XPS has readahead, you may want to turn that up too. I'd probably start at reading ahead by a second or two and adjust from there. If that doesn't help, you may be having trouble consuming the I/O produced. You'll need to look for ways to improve parallelism at that point, but my guess is that raw I/O bandwidth is your bottleneck. Greg Doug Johnson wrote: > > Question: > Have you ever seen your system run XPS simple table scans at higher than > 5-10 I/O MB per CPU second? > snip..... > Have you ever seen XPS run scans (or anything else) at higher than 5-10 > I/O MB per CPU second on any Unix (or NT) system or is this the current > state of equilibrium? Thanks, > > Doug