I have a module with approximately 8,000 records in it. I have filtered record lists for this to bring the return down to approximately 1700 records, and I’ve kept the return to four fields, and the paging down to 25 records per page, but the load time is still unacceptably slow.
Do you have any suggestions to improve performance? This is a reference table for positions in our organization - it will be looked at often.
Based on the information you provided it could be due to hardware. We have a similar case 20k records filtered down to 9k including a like filter Email not like '%foobar.com', which is even more expensive. 50 records per page takes around 3 seconds.
Can you provide your timing data?
8K records → filtered to 1388 records → displaying 100 records per page = 3 min 10 seconds to display first page. 2 min 3 seconds to display second page of 100 records.
Prefilter is one field = ‘value’. Nothing fancy or overly resource intensive.
This module relates to several other modules, but at this time this is the only one of those modules that has any data in it.
A ping sent to the server returned this:
Pinging 162.251.146.168 with 32 bytes of data:
Reply from 162.251.146.168: bytes=32 time=212ms TTL=54
Reply from 162.251.146.168: bytes=32 time=70ms TTL=54
Reply from 162.251.146.168: bytes=32 time=112ms TTL=54
Reply from 162.251.146.168: bytes=32 time=111ms TTL=54
Ping statistics for 162.251.146.168:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 70ms, Maximum = 212ms, Average = 126ms
Seems reasonable, so I don’t think it’s a connection issue.
Is there a cache that might be getting full?
OK, I played around with the virtual server settings. Increased the server memory from 2GB to 8GB and brought the query response time down to about 6 seconds. Looks like that was the bottleneck.
A bloated database table (separate issue) may also have contributed to the slowdown. After solving that issue, along with beefing up the server, things are running much faster.