cache warmer?

[#1]

Happy M-day, everyone in the USA.

So, I've notice one bit of thing that is helping the slow browsing issue.  When I open my dashboard from the browser, for the first time, it is slow.  using a different browser, later, turn out to be faster.

It reminds me of the MS product that, they have a way to load the data to be browse into memory.  Someone call that technique a cache warming; liked the data is cached and all warm-up ready to go.

is there anything liked that for QV11?

I have a App server, web server, and a Dist server, then clients.

[Clever_Anjos]

"instantly " = a few seconds, yes

[#3]

good point.

pre load- check

manually reload - check.

slow.

[#4]

Thank you, swuehl !

A lot of good reads.

[Peter_Cammaert]

If I read your post correctly, you may be experiencing the simple load of a document in memory by QVS. The very frist user that opens a document will cause the QVS to load the document in memory. That in itself will account for a delay from a few seconds to many minutes, depending on the size of your document. The solution to this, as explained by the others, is the pre-load option which causes the QVS to load a document in memory immediately after a restart and whenever a document is reloaded. Even when no-one really needs or uses that document at that moment.

This is especially useful for the larger documents, as nobody likes to wait a quarter of an hour or more for those hundreds of gigs to load. And especially not when you can be sure that after a period of disuse, these documents may get kicked out of server RAM again.

After loading the document in memory (no cache) various calculations and aggregations wil create subsets of data that will be cached for real. Reading the same dataset and esxpression results from the cache instead of having to recalculate will speed up selections and object filling.

All of this impacts both RAM usage and response speed.

Best,

Peter

[alexdataiq]

So, does the preload option indeed helps to seed the cache of document?

Cheers

[oknotsen]

"slow"? What slow, who slow, why slow? You might want to be a little bit more descriptive .

[Peter_Cammaert]

AFAIK no. Pre-load doesn't simulate users, It just loads the document in RAM before anyone uses it.

[oknotsen]

The document has no cache; the server does.

Preload only loads the document in memory. This can already be a time saver for the first person to use the document each day as huge documents can take some time.

Calculations in cache is only stored if actual selections are made. What you can do is "cache pre-heating" by for example run a "test script" that makes common selections after the document is reloaded.

[#10]

may be it is fast for some apps.  the app, i've adopted here, isn't getting noticeable speed.

[oknotsen]

Preload will only be noticable for documents of some size.

I had a customer with 2 billion rows of data in his apps. The difference was from 5+ minutes to just seconds.